Agent Skills: start-workflow

Codex compatibility note:

• Invoke repository skills with $skill-name in Codex; this mirrored copy rewrites legacy Claude /skill-name references.
• Task tracker mandate: BEFORE executing any workflow or skill step, create/update task tracking for all steps and keep it synchronized as progress changes.
• User-question prompts mean to ask the user directly in Codex.
• Ignore Claude-specific mode-switch instructions when they appear.
• Strict execution contract: when a user explicitly invokes a skill, execute that skill protocol as written.
• Subagent authorization: when a skill is user-invoked or AI-detected and its protocol requires subagents, that skill activation authorizes use of the required spawn_agent subagent(s) for that task.
• Do not skip, reorder, or merge protocol steps unless the user explicitly approves the deviation first.
• For workflow skills, execute each listed child-skill step explicitly and report step-by-step evidence.
• If a required step/tool cannot run in this environment, stop and ask the user before adapting.

Codex Project-Reference Loading (No Hooks)

Codex uses static project-reference loading instead of runtime-injected project docs. When coding, planning, debugging, testing, or reviewing, open project docs explicitly using this routing.

Always read:

• docs/project-config.json (project-specific paths, commands, modules, and workflow/test settings)
• docs/project-reference/docs-index-reference.md (routes to the full docs/project-reference/* catalog)
• docs/project-reference/lessons.md (always-on guardrails and anti-patterns)

Missing/stale context route: If docs/project-config.json, the docs index, lessons.md, CLAUDE.md, AGENTS.md, or any task-required reference doc is missing or stale, auto-run $project-init or the narrow setup route ($project-config, $docs-init, $scan-all, $scan --target=<key>, $claude-md-init) before ordinary project-specific work. If Codex mirrors or AGENTS.md are missing/stale, ask the user to run $sync-codex; do not auto-run it.

Situation-based docs:

• Backend/CQRS/API/domain/entity changes: backend-patterns-reference.md, domain-entities-reference.md, project-structure-reference.md
• Frontend/UI/styling/design-system: frontend-patterns-reference.md, scss-styling-guide.md, design-system/README.md
• Spec authoring, docs/specs/ pathing, or TC format: feature-spec-reference.md, spec-system-reference.md, spec-principles.md
• Behavior/public-contract changes or spec-test-code sync: workflow-spec-test-code-cycle-reference.md plus the spec docs above
• Derived spec indexes/ERDs/reimplementation guides: spec-system-reference.md and source Feature Specs under docs/specs/
• Integration test implementation/review: integration-test-reference.md
• E2E test implementation/review: e2e-test-reference.md
• Code review/audit work: code-review-rules.md plus domain docs above based on changed files

Do not read all docs blindly. Start from docs-index-reference.md, then open only relevant files for the task.

Quick Summary

Goal: Detect user intent → auto-select direct/skill/standard/custom path → activate with full task tracking plan.

Workflow:

1. Detect — Execute explicit /workflow-* or $start-workflow <id> directly; otherwise match prompt against workflow catalog and skill list
2. Auto-select — Choose direct execution, a skill, a standard workflow, or a custom pipeline without asking the user to pick the path
3. Activate — Create ALL task tracking items for chosen sequence; mark first in_progress

Key Rules:

• MUST ATTENTION define success criteria before execution and loop until observable verification passes.

• MUST ATTENTION when creating/reviewing specs or tests, name Business Intent / Invariant Guarded or the protected business intent/invariant and ensure the test would fail if that intent breaks.

• MUST ATTENTION auto-select the best execution path for ordinary prompts. Do not ask the user to choose between direct execution, skill, standard workflow, or custom workflow.

• Explicit /workflow-* or $start-workflow <id> invocation counts as the user choosing that workflow; execute it directly.

• Propose Custom Pipeline when no catalog workflow is a strong fit (>80% steps relevant = use catalog)

• workflows.json workflows field is an OBJECT — use workflows[workflowId], NEVER .find() or [index]

• Create ALL task tracking items BEFORE marking the first task in_progress — batch creation, then execute

• NEVER mark a task completed without invoking its skill invocation — skip = in_progress + comment, not delete

• ALWAYS check context for ## Workflow Catalog first (Tier 1) — NEVER read workflows.json directly when catalog is in context

• If another workflow is active, it auto-switches (ends current, starts new) — no manual cleanup needed

NOT for: Manual step execution (follow task tracking items), workflow design (use plan), catalog management.

Related: $start-workflow <workflowId> | Catalog: static ## Workflow Catalog baked into CLAUDE.md (no router/tracker hooks)

Custom Pipeline Option

When the prompt doesn't cleanly match a single catalog workflow — or combining steps from multiple workflows serves the request better — the AI MAY propose a Custom Pipeline alongside the catalog option.

When to propose

| Condition | Example | | -------------------------------------------- | ------------------------------------------------------------------------------------ | | No catalog workflow matches well | "Review hook changes and update skill docs" — spans review + docs | | Best-match has significant unnecessary steps | Quick investigate + fix, but workflow-bugfix includes full TDD + integration cycle | | Prompt combines 2+ workflow domains | "Audit performance and write integration tests for the slow query" | | User explicitly requests a step sequence | "Just run scout, plan, and feature-implement — nothing else" |

Do NOT propose when a catalog workflow is a strong match (>80% of its steps are relevant). Catalog workflows encode validated best-practice sequences — prefer them.

How to build

1. Valid steps only — Use only canonical step ids — those appearing in workflow sequence arrays in workflows.json (each maps to a real .claude/skills/<step>/SKILL.md and is invoked as /<step>). No invented step names.
2. Logical order — Investigate → Plan → Implement → Test. Never reverse dependency order.
3. Minimal — Include only steps the prompt needs. No "just in case" additions.
4. Name it — Short descriptive name: "Quick Fix + Docs", "Audit + Test Coverage".

How to present (ask the user directly format)

Show full step sequences for ALL options so the user compares scope:

Option A — Activate "Bug Fix" workflow (Recommended)
  Steps: $scout → $investigate → $debug-investigate → $plan → $fix → $prove-fix → $test → ...

Option B — Custom Pipeline: "Quick Fix + Docs"
  Steps: $scout → $investigate → $fix → $docs-update
  Rationale: Prompt targets a known location — full TDD cycle is over-engineered here.

Rules:

• Always show full step list per option
• One-sentence AI rationale for the custom pipeline
• Catalog workflow = "(Recommended)" unless custom pipeline confidence is clearly higher
• NEVER present custom pipeline as the only option — always include the catalog option
• For project-specific architecture, test, documentation, naming, or workflow rules, read docs/project-config.json and docs/project-reference/docs-index-reference.md; keep this reusable start-workflow protocol generic.

Task creation for Custom Pipeline

Same 1:1 protocol — one task tracking per step. Use [Custom] prefix to distinguish from catalog tasks:

Task tracking: subject="[Custom] /{step-name} — {brief description}", description="Custom pipeline step N/{total}.", activeForm="Executing /{step-name}"

Workflow Lookup — Token-Efficient (3-Tier Strategy)

ALWAYS try tiers in order — stop at first success.

Tier 1: Context (FREE — no file reads)

The workflow catalog is already present as the ## Workflow Catalog section in your context — baked statically into CLAUDE.md (and the AGENTS.md mirror), not injected by any hook.

1. Search your context for ## Workflow Catalog
2. Find the line: **{workflowId}** — {name}
3. Read the NEXT line: Use: ... | Steps: /step1 → /step2 → ...
4. Parse the Steps: value — these ARE the slash commands for task tracking

Example: Steps: $scout → $plan → $feature-implement → $test → $workflow-end → Create 5 task tracking items for $scout, $plan, $feature-implement, $test, $workflow-end in order.

✅ Use Tier 1 for: all standard task tracking creation (no file reads needed) ⚠️ Use Tier 2 if: catalog not in context, OR you need preActions.injectContext

Tier 2: Targeted Grep (1 grep operation)

Use when catalog is absent from context OR preActions.injectContext is needed:

Grep: pattern='"<workflowId>":'  path='.claude/workflows.json'  context=35

Returns only that workflow's entry (~35 lines vs full file). Parse: sequence array → step IDs → invoke each as /<stepId>.

Tier 3: Minimal Read (last resort — avoid)

Only if Tier 1 and Tier 2 both fail:

Grep '"<workflowId>":' context=30    ← that workflow's entry only

NEVER read the full file — it is large and wastes tokens.

After Activation — Task Creation Protocol (ZERO TOLERANCE)

Active-goal resolution (BEFORE child task creation): resolve the active Goal Contract per SYNC:goal-contract-satisfaction-loop — active plan goal.md, else plans/goals/{YYMMDD-HHmm}-{slug}/goal.md, else create one from the current user request using .claude/templates/goal-contract-template.md. Record the resolved goal path and pass it to every child step/sub-agent so the whole workflow executes against the same saved success criteria. The workflow may end only when the goal's Goal Satisfaction matrix passes or a blocker is escalated.

FIRST action after activation: create EXACTLY one task tracking for EACH entry in the workflow's sequence array.

How to read workflows.json — CRITICAL SCHEMA

workflows.json is a JSON OBJECT, not an array. Most common AI mistake.

{
  "settings":       { ... },
  "workflows":      { <workflowId>: WorkflowEntry }   ← OBJECT, keyed by ID
}

Lookup algorithm:

workflow = workflows[workflowId]           // key lookup — NOT .find(), NOT [index]
steps    = workflow.sequence               // array of step ID strings
// resolve slash command — step id IS the command:
slashCmd = "/" + stepId                     // "scout" → "$scout"

WorkflowEntry fields:

| Field | Type | Notes | | ------------ | -------- | --------------------------------------- | | name | string | Display name | | sequence | string[] | Ordered step IDs — SOLE source of truth | | whenToUse | string | Natural language intent matching | | preActions | object | Optional injectContext / readFiles |

FORBIDDEN (common mistakes):

// ❌ WRONG
workflows.find(w => w.id === workflowId)
workflows[0]

// ✅ CORRECT
workflows[workflowId]
Object.keys(workflows)   // list all IDs

Task creation steps

1. Tier 1 first (no file read): Search context for ## Workflow Catalog → find **{workflowId}** → parse Steps: line → slash commands are ready to use
2. Tier 2 if needed: Grep .claude/workflows.json --pattern '"<workflowId>":' --context 35 → parse sequence, invoke each step as /<stepId>
3. Create one task tracking per step IN ORDER

See Workflow Lookup — Token-Efficient (3-Tier Strategy) above for full lookup rules and fallback chain.

Task format:

Task tracking: subject="[Workflow] /{step-name} — {brief description}", description="Workflow step N/{total}. {conditional note}", activeForm="Executing /{step-name}"

Rules (NON-NEGOTIABLE):

• 1:1 mapping — each sequence entry = exactly one task. No consolidation, no invented tasks.
• Conditional steps still get tasks — add to description: "Conditional — skip if reviews pass"
• Recursive self-calls get tasks — e.g., [Workflow] $workflow-review-changes — Recursive re-review (conditional)
• Count verification — after creation: task count == len(sequence). Fix mismatch before proceeding.

Create ALL tasks first → then TaskUpdate first task to in_progress.

Step Execution Protocol

Per step: TaskUpdate in_progress → invoke skill invocation → complete skill → TaskUpdate completed.

• Completing a task without invoking its skill invocation = workflow violation
• Validation gates ($plan-validate, $plan-review, $why-review) MUST use explicit evidence and local project protocol — NEVER auto-approve inferred decisions. Explicit user approval in the prompt may satisfy the gate only when the gate's skill permits it.
• To skip a conditional step: TaskUpdate in_progress → comment "Skipped — {reason}" → TaskUpdate completed. Never delete.

Workflow-in-Workflow Gate (HARD GATE — NO EXCEPTIONS)

Some workflow steps ARE themselves full workflows. Running them inline causes the parent session to absorb the entire nested workflow's tool calls, file reads, and sub-agent reports — guaranteed context overflow on long sequences.

Steps requiring sub-agent delegation (hard gate):

| Step | Workflow activated | Step count source | Agent type | | -------------------------- | ------------------------- | ---------------------------------------------------- | --------------- | | $workflow-review-changes | workflow-review-changes | len(workflows["workflow-review-changes"].sequence) | code-reviewer |

Protocol when these steps appear in the active workflow sequence:

1. NEVER invoke via inline skill invocation call
2. Spawn via spawn_agent tool: agent_type: "code-reviewer"
3. Agent prompt must include: current git diff context + feature/task description
4. Sub-agent runs the full nested workflow in its isolated context
5. Return ONLY SYNC:subagent-return-contract summary — write full findings to plans/reports/
6. Main agent reads plans/reports/ file only when resolving specific blockers

The ⚠️ [WORKFLOW-IN-WORKFLOW GATE] is model-driven: apply it yourself whenever the next step is one of the above — no hook emits this warning.

IMPORTANT MANDATORY Steps: detect-workflow -> analyze-best-match -> auto-select-execution-path -> activate-workflow -> create-task-tracking -> execute-sequence

IMPORTANT MANDATORY Steps: detect-workflow -> analyze-best-match -> auto-select-execution-path -> activate-workflow -> create-task-tracking -> execute-sequence

[MANDATORY] task tracking FIRST — break every workflow into tasks before any action. NEVER skip. [MANDATORY] Auto-select the best path for auto-detected workflows; do not use a direct user question for workflow-selection confirmation. Explicit workflow invocation executes directly. [MANDATORY] skill invocation REQUIRED per step — NEVER mark a task completed without invoking it.

AI Mistake Prevention — Failure modes to avoid on every task:

Re-read files after context changes. Context compaction, resume, or long-running work can make memory stale; verify current files before acting. Verify generated content against source evidence. AI hallucinates APIs, names, claims, and document facts. Check the relevant source before documenting or referencing. Check downstream references before deleting or renaming. Removing an artifact can stale docs, generated mirrors, configs, and callers; map references first. Trace the full impact chain after edits. Changing a definition can miss derived outputs and consumers. Follow the affected chain before declaring done. Verify ALL affected outputs, not just the first. One green check is not all green checks; validate every output surface the change can affect. Assume existing values are intentional — ask WHY before changing. Before changing a constant, limit, flag, wording, or pattern, read nearby context and history. Surface ambiguity before acting — don't pick silently. Multiple valid interpretations require an explicit question or stated assumption with risk. Keep shared guidance role-relevant. Universal guidance must help every receiving skill or agent; code-specific obligations belong only in code-specific protocols.

Critical Thinking Mindset — Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act. Anti-hallucination: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination.

Incremental Result Persistence — MANDATORY for all sub-agents or heavy inline steps processing >3 files.

1. Before starting: Create report file plans/reports/{skill}-{date}-{slug}.md
2. After each file/section reviewed: Append findings to report immediately — never hold in memory
3. Return to main agent: Summary only (per SYNC:subagent-return-contract) with Full report: path
4. Main agent: Reads report file only when resolving specific blockers

Why: Context cutoff mid-execution loses ALL in-memory findings. Each disk write survives compaction. Partial results are better than no results.

Report naming: plans/reports/{skill-name}-{YYMMDD}-{HHmm}-{slug}.md

Sub-Agent Return Contract — When this skill spawns a sub-agent, the sub-agent MUST return ONLY this structure. Main agent reads only this summary — NEVER requests full sub-agent output inline.

## Sub-Agent Result: [skill-name]

Status: ✅ PASS | ⚠️ PARTIAL | ❌ FAIL
Confidence: [0-100]%

### Findings (Critical/High only — max 10 bullets)

- [severity] [file:line] [finding]

### Actions Taken

- [file changed] [what changed]

### Blockers (if any)

- [blocker description]

Full report: plans/reports/[skill-name]-[date]-[slug].md

Main agent reads Full report file ONLY when: (a) resolving a specific blocker, or (b) building a fix plan. Sub-agent writes full report incrementally (per SYNC:incremental-persistence) — not held in memory.

Context budget — the return payload is a SUMMARY, not a transcript: ≤10 finding bullets, no raw file contents / full diffs / verbatim logs inline, no re-pasted source. Everything beyond the summary lives in the Full report on disk. A sub-agent that would exceed the summary shape MUST write the detail to its report and return only the pointer — the orchestrator's context is the scarce resource the whole map-reduce protects.

MUST ATTENTION apply critical + sequential thinking — every claim needs appropriate traced evidence (file:line for repo/code claims; source URL or artifact section for research, product, content, and docs claims); confidence >80% to act, <60% DO NOT recommend. Anti-hallucination: never present guess as fact, admit uncertainty freely, cross-reference independently, stay skeptical of own confidence.

MUST ATTENTION apply AI mistake prevention — verify generated content against evidence, trace downstream references before deleting or renaming, verify all affected outputs, re-read files after context loss, and surface ambiguity before acting.

• MANDATORY Resolve the active Goal Contract BEFORE work (active plan goal.md → plans/goals/{YYMMDD-HHmm}-{slug}/goal.md → create from current request) and read saved success criteria before editing.
• MANDATORY Append iteration evidence after execution; emit a Goal Satisfaction matrix (PASS/FAIL/BLOCKED) before reporting PASS; loop on validated FAIL; escalate repeated no-progress or blockers. NEVER store secrets in goal files.

Closing Reminders

Protocols in force (concise digest of the SYNC/shared blocks this skill carries):

• AI Mistake Prevention: verify generated content against evidence, trace downstream references, verify all affected outputs, re-read after context loss, surface ambiguity.
• Critical Thinking: traced file:line proof, confidence >80%; NEVER present guess as fact.
• Incremental Persistence: append findings to report per file; NEVER hold in memory.
• Sub-Agent Return Contract: sub-agents return summary only; NEVER inline full output.

MUST ATTENTION auto-select the best path for ordinary prompts; explicit /workflow-* or $start-workflow <id> invocation executes directly. Do not ask for workflow-selection confirmation. MUST ATTENTION workflows is an OBJECT — workflows[workflowId], NEVER .find() / [index] / .forEach() MUST ATTENTION create ALL task tracking items for the full sequence BEFORE marking the first task in_progress MUST ATTENTION never mark a task completed without invoking its skill invocation — skip means comment + completed, not delete MUST ATTENTION custom pipeline steps must be canonical step ids (each maps to a real .claude/skills/<step>/SKILL.md) — never invent step names MUST ATTENTION use Tier 1 context parse FIRST — check ## Workflow Catalog in context before any file read

[TASK-PLANNING] Before acting, analyze task scope and systematically break it into small todo tasks and sub-tasks using task tracking.

[IMPORTANT] Analyze how big the task is and break it into many small todo tasks systematically before starting — this is very important.

Hookless Prompt Protocol Mirror (Auto-Synced)

Source: .claude/.ck.json + .claude/skills/shared/sync-inline-versions.md (:full blocks) + .claude/scripts/lib/hookless-prompt-protocol.cjs

[WORKFLOW-EXECUTION-PROTOCOL] [BLOCKING] Workflow Execution Protocol — MANDATORY IMPORTANT MUST CRITICAL. Do not skip for any reason.

Generic portability boundary: Reusable skills and protocol text stay project-neutral; project-specific conventions are discovered from docs/project-config.json and docs/project-reference/. Apply shared AI-SDD from shared/sdd-artifact-contract.md. Read docs/project-config.json and docs/project-reference/docs-index-reference.md, then open the project reference docs named there. For spec, test-case, behavior-change, public-contract, or docs/specs/ work, route through the local spec docs named by the docs index: feature-spec-reference.md, spec-system-reference.md, spec-principles.md, and workflow-spec-test-code-cycle-reference.md when specs/tests/code must stay synchronized. If either file or a required reference doc is missing or stale, auto-run $project-init (or the narrow lower-level route such as $project-config, $docs-init, $scan-all, or $scan --target=<key>) before ordinary project-specific work. Any supported AI tool may execute when this shared context and local docs are available.

1. DETECT: If the prompt starts with an explicit slash skill/workflow command, execute it directly. Otherwise match the prompt against the workflow catalog and skill list.
2. ANALYZE: Choose the best option: execute directly, invoke a skill, activate a standard workflow, or compose a custom step combination.
3. AUTO-SELECT: Pick the best option yourself. Do not ask the user to choose between direct execution, skill, standard workflow, or custom workflow.
4. ACTIVATE: For a selected workflow, call $start-workflow <workflowId>; for a selected skill, invoke that skill; for a custom workflow, sequence custom steps directly; for direct execution, proceed with the task.
5. CREATE TASKS: task tracking for ALL workflow/skill/custom steps before execution when the selected path has multiple steps.
6. EXECUTE: Advance per the Workflow Step Advancement & Parallel Phases rule in your context instructions — model-driven; a sub-agent completion advances a step identically to an inline call; a parallel-phase group is an all-return barrier (advance only after ALL members return, never serialize it)

Shared AI-SDD Protocol Markers

Source: .claude/skills/shared/sync-inline-versions.md

SYNC:ai-sdd-artifact-contract

AI-SDD Artifact Contract — Shared spec-driven development rules stay portable and source-owned.

1. Keep reusable AI-SDD principles in .claude; put repository-specific paths, commands, owners, products, and formats in project config/reference docs.
2. Preserve cycle: spec -> plan -> tasks -> implement -> verify -> update spec/docs.
3. Trace every requirement or invariant through decision, task, TC/test, source evidence, and docs/spec update.
4. Treat code-to-spec extraction as reference-only until accepted by the canonical spec owner.
5. Any supported AI tool may plan, implement, review, or verify with synced context; using multiple tools is optional.
6. Update .claude source first, then sync generated mirrors; do not manually edit .agents, .codex, or AGENTS.md. — why: mirrors are generated artifacts; hand-edits are overwritten on the next sync
7. If docs/project-config.json, root instruction files, or a required project-reference doc is missing or stale, auto-run $project-init or the narrow lower-level route before ordinary project-specific work.

Active reference: shared/sdd-artifact-contract.md in the active skills root.

SYNC:ai-sdd-artifact-contract:reminder

• MANDATORY Apply shared/sdd-artifact-contract.md; keep reusable AI-SDD in .claude and local rules in project docs.
• MANDATORY Code-to-spec extraction is reference-only until canonical acceptance; any supported AI tool may execute with synced context.
• MANDATORY Update .claude source before syncing generated mirrors; do not manually edit .agents, .codex, or AGENTS.md.
• MANDATORY Missing or stale project config, root instruction files, or required reference docs route project-specific work through $project-init or the narrow setup route automatically. [TASK-PLANNING] [MANDATORY] BEFORE executing any workflow or skill step, create/update task tracking for all planned steps, then keep it synchronized as each step starts/completes.

[LESSON-LEARNED-REMINDER] [BLOCKING] Task Planning & Continuous Improvement — MANDATORY. Do not skip.

Break work into small tasks (task tracking) before starting. Add final task: "Analyze AI mistakes & lessons learned".

Extract lessons — ROOT CAUSE ONLY, not symptom fixes:

1. Name the FAILURE MODE (reasoning/assumption failure), not symptom — "assumed API existed without reading source" not "used wrong enum value".
2. Generality test: does this failure mode apply to ≥3 contexts/codebases? If not, abstract one level up.
3. Write as a universal rule — strip project-specific names/paths/classes. Useful on any codebase.
4. Consolidate: multiple mistakes sharing one failure mode → ONE lesson.
5. Recurrence gate: "Would this recur in future session WITHOUT this reminder?" — No → skip $learn.
6. Auto-fix gate: "Could $code-review/$code-simplifier/$security-review/$lint catch this?" — Yes → improve review skill instead.
7. BOTH gates pass → ask user to run $learn. [CRITICAL-THINKING-MINDSET] Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act. Anti-hallucination principle: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination. AI Attention principle (Primacy-Recency): Put the 3 most critical rules at both top and bottom of long prompts/protocols so instruction adherence survives long context windows. Goal-driven execution: Define success criteria first, loop until verified, and stop only when observable checks pass. Tests verify intent: Tests must protect business rules/invariants and fail when the protected intent breaks, not only mirror current behavior.

Common AI Mistake Prevention (System Lessons)

• Re-read files after context compaction. Edit requires prior Read in same context; compaction wipes read state. Re-read before editing.
• Grep for old terms after bulk replacements. AI over-trusts find/replace completeness. Grep full repo after bulk edits for missed refs in docs/configs/catalogs.
• Check downstream references before deleting. Deletions cascade doc/code staleness. Map referencing files before removal.
• After memory loss, check existing state before creating new. Compaction wipes prior-work memory. Query current state to resume — never blindly duplicate.
• Verify AI-generated content against actual code. AI hallucinates APIs, class names, method signatures. Grep to confirm existence before documenting/referencing.
• Trace full dependency chain after edits. Changing a definition misses downstream consumers. Trace the full chain.
• When renaming, grep ALL consumer file types. Some file types silently ignore missing refs (no compile error). Search code, templates, configs, generated files.
• Trace ALL code paths when verifying correctness. Code existing ≠ code executing. Trace early exits, error branches, conditional skips — not just happy path.
• Update docs that embed canonical data when source changes. Docs inlining derived data (workflows, schemas, configs) go stale silently. Update all embedding docs alongside source.
• Verify sub-agent results after context recovery. Background agents may finish while parent compacted — grep-verify output, don't trust assumed completion.
• Cross-check full target list against sub-agent assignments. Parallel sub-agents by category miss boundary items. Reconcile union of assignments against target list before proceeding.
• Sub-agents inherit knowledge only from their agent .md definition — use custom agent types, not built-in Explore. Tool adoption = permission + knowledge + enforcement (numbered workflow step).
• Persist sub-agent findings incrementally, not as a final batch. Long sub-agents hit cutoffs before final write — findings lost. Instruct append-per-section to report file.
• When debugging, ask "whose responsibility?" before fixing. Trace caller (wrong data) vs callee (wrong handling). Fix at responsible layer — never patch symptom site.
• Grep ALL removed names after extraction/refactoring. Primary file "done" ≠ secondary files clean. Grep entire scope for every removed symbol before declaring complete.
• Assume existing values are intentional — ask WHY before changing. Pattern-matching as "wrong" skips context. Before changing any constant/limit/flag: read comments, git blame, surrounding code.
• Verify ALL affected outputs, not just the first. One build green ≠ all green. Multi-stack changes (backend/frontend/tests/docs) require verifying EVERY output.
• Evaluate fit before copying a nearby pattern. Closest example ≠ matching preconditions — verify the new context shares the same constraints, base classes, scope, lifetime.
• Holistic-first debugging — resist nearest-attention trap. Don't dive into first plausible cause. List EVERY precondition (config, env vars, paths, DB, endpoints, creds, versions, DI, data). Verify each against evidence (grep/query — not reasoning). Ask "what would falsify this?" — if nothing, it's not a hypothesis. Most expensive failure: going deeper in "obvious" layer while bug sits in layer never questioned.
• Surgical changes — apply the diff test (context-aware). Two modes: (1) Bug fix → every line traces to the bug; no restyling; orphan cleanup only for imports YOUR changes made unused. (2) Review/enhancement → implement improvements AND announce as "Enhancement beyond main request: [what]". Never silently scope-creep. Diff test: "Would this line exist if I wasn't asked to do X?" — if no, delete or announce.
• Surface ambiguity before coding — don't pick silently. Multiple valid interpretations → present each with effort: "[Request] could mean (1) [N h], (2) [N h]. Which matters?" List scope/format/volume/constraints assumptions first. If simpler path exists, say so. Never silently pick.
• [MANDATORY FIRST ACTION] ALWAYS activate a suitable skill or workflow BEFORE responding. Match task against workflow catalog + skill list; invoke via skill invocation or $start-workflow <workflowId>. NEVER answer or write code before checking. Skip = protocol violation.
• Why-Review adversarial mindset — apply when reviewing any plan, decision, or design. Default SKEPTIC not VALIDATOR: steel-man a rejected alternative, invert each stated reason ("what does it sacrifice?"), stress-test top 2-3 assumptions, run pre-mortem ("ships, fails in 3 months — what breaks?"), surface 1-2 alternatives author missed. Section presence ≠ quality; quality = causal reasoning + concrete mitigations + evidence, not "it's better" or "monitor closely".
• Front-load report-write in sub-agent prompts for large reviews. Many-file sub-agents hit budget before final write — findings lost. Design prompts so: (1) report-write is first explicit deliverable, (2) append per-file/section (not batched), (3) scope bounded so reads don't exhaust budget. Truncated mid-sentence with no report file → spawn narrower scope, don't retry same prompt.
• After context compaction, re-verify all prior phase outcomes before continuing. Summaries describe intent, not environment state (git index, filesystem, processes). On resume, FIRST audit: git status, re-read modified files, verify filesystem. Every "completed" claim is an untested hypothesis until evidence confirms.
• OOM/memory: check row count before row size. Triage: (1) Unbounded query — no DB filter for trigger? Push filter to DB; eliminates OOM. (2) Large rows? Projection reduces proportionally. Row reduction > projection in ROI.
• Keep domain concepts out of generic/shared/infrastructure layers. Reusable layer (shared library, framework, infra module) must reference NO consumer-specific domain concept — tenant/customer/product IDs, business entities, feature rules. Leak compiles + runs → passes review silently while coupling the "reusable" layer to one consumer. Keep shared type domain-free; push domain fields/logic down into the consumer via subclass/composition. — why: a layer coupled to one consumer's domain is no longer reusable.