tech-stack-research Skill

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.

[BLOCKING] Execute skill steps in declared order. NEVER skip, reorder, or merge steps without explicit user approval. [BLOCKING] Before each step or sub-skill call, update task tracking: set in_progress when step starts, set completed when step ends. [BLOCKING] Every completed/skipped step MUST include brief evidence or explicit skip reason. [BLOCKING] If Task tools are unavailable, create and maintain an equivalent step-by-step plan tracker with the same status transitions.

Quick Summary

Goal: Deliver a user-confirmed, per-layer tech stack — each choice backed by 3+ researched options, weighted scoring, cited evidence, and a confidence % — by acting as a solution architect who derives technical requirements from business analysis, researches the current market, and produces a detailed comparison report, so the team commits to a stack fit for scale, budget, skills, and timeline, not familiarity.

Summary:

Requirements come BEFORE research: load prior business/domain/PBI artifacts, map business signals to technical requirements, and gate on user confirmation (a direct user question) before any WebSearch.
Evaluate every stack layer (backend, frontend, database, messaging, infra, auth) independently — minimum 3 WebSearched options per layer, each with cited evidence (URL, benchmark, case study), never familiarity.
Score with the weighted 8-criteria matrix (High=3x / Medium=2x / Low=1x), then rank each layer with a confidence %; capped <=200-line report goes to {plan-dir}/research/tech-stack-comparison.md.
The end-of-skill user validation interview (5-8 questions) is mandatory and never skipped — only confirmed decisions get written to phase-02-tech-stack.md as status: confirmed.

Workflow:

Load Business Context — Read business evaluation, domain model, refined PBI artifacts
Derive Technical Requirements — Map business needs to technical constraints
Research Per Layer — WebSearch top 3 options for each stack responsibility
Deep Compare — Pros/cons matrix, benchmarks, community health, team fit
Score & Rank — Weighted scoring across 8 criteria
Generate Report — Structured comparison report with recommendation
User Validation — Present findings, ask 5-8 questions, confirm choices

Key Rules:

MANDATORY IMPORTANT MUST ATTENTION research minimum 3 options per stack layer
MANDATORY IMPORTANT MUST ATTENTION include confidence % with evidence for every recommendation
MANDATORY IMPORTANT MUST ATTENTION run user validation interview at end (never skip)
All claims must cite sources (URL, benchmark, case study)
Recommend based on benchmarked evidence (URL, benchmark, case study); never on familiarity alone

Be skeptical. Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence percentages (Idea should be more than 80%).

Step 1: Load Business Context

Read artifacts from prior workflow steps (search in plans/ and team-artifacts/):

Business evaluation report (viability, scale, constraints)
Domain model / ERD (complexity, entity count, relationships)
Refined PBI (acceptance criteria, scope)
Discovery interview notes (team skills, budget, timeline)

Extract and summarize:

| Signal | Value | Source | | ---------------------- | ------------ | ------------------- | | Expected users | ... | discovery interview | | Domain complexity | Low/Med/High | domain model | | Team skills | ... | discovery interview | | Budget constraint | ... | business evaluation | | Timeline | ... | business evaluation | | Compliance needs | ... | business evaluation | | Real-time needs | Yes/No | refined PBI | | Integration complexity | Low/Med/High | domain model |

Step 2: Derive Technical Requirements

Map business signals to technical requirements:

| Business Signal | Technical Requirement | Priority | | ------------------ | ----------------------------------------------- | -------- | | High user scale | Horizontal scaling, connection pooling | Must | | Complex domain | Strong type system, ORM with migrations | Must | | Real-time features | WebSocket/SSE support, event-driven arch | Must | | Small team | Low learning curve, good DX, batteries-included | Should | | Tight budget | Open-source, low hosting cost | Should | | Compliance | Audit trail, encryption, auth framework | Must |

MANDATORY IMPORTANT MUST ATTENTION validate derived requirements with user via a direct user question before proceeding to research.

Step 3: Research Per Stack Layer

For EACH layer, research top 3 options using WebSearch (minimum 5 queries total):

Stack Layers to Evaluate

| Layer | Example Options | Research Focus | | ---------------------- | ------------------------------------- | ------------------------------------- | | Backend Framework | Candidate backend runtimes/frameworks | Performance, type safety, ecosystem | | Frontend Framework | Candidate frontend frameworks | DX, ecosystem, hiring, enterprise fit | | Database | Candidate database engines/stores | Scale, query complexity, cost | | Messaging/Events | Candidate messaging/event systems | Throughput, reliability, complexity | | Infrastructure | Docker+K8s, Serverless, PaaS | Cost, ops overhead, scaling | | Auth | Keycloak, Auth0, custom | Cost, compliance, flexibility |

WebSearch Queries (minimum 5 per layer)

"{option_A} vs {option_B} {current_year} comparison"
"{option} enterprise production case studies"
"{option} community size github stars"
"{option} performance benchmarks {use_case}"
"{option} security track record vulnerabilities"

Step 4: Deep Comparison Matrix

For EACH stack layer, produce a comparison table:

| Criteria | Option A | Option B | Option C | Weight | | -------------------- | ----------------- | -------- | -------- | ------ | | Team Fit | score + rationale | ... | ... | High | | Scalability | score + rationale | ... | ... | High | | Time-to-Market | score + rationale | ... | ... | High | | Ecosystem/Libs | score + rationale | ... | ... | Medium | | Hiring Market | score + rationale | ... | ... | Medium | | Cost (hosting) | score + rationale | ... | ... | Medium | | Learning Curve | score + rationale | ... | ... | Medium | | Community Health | score + rationale | ... | ... | Low |

Scoring: 1-5 scale. Weight: High=3x, Medium=2x, Low=1x.

Per-Option Detail Block

For each option, document:

### {Layer}: {Option Name}

**Pros:**

- {Pro 1} — {evidence/source}
- {Pro 2} — {evidence/source}
- {Pro 3} — {evidence/source}

**Cons:**

- {Con 1} — {evidence/source}
- {Con 2} — {evidence/source}

**Best suited when:** {conditions}
**Not suitable when:** {conditions}
**Production examples:** {2-3 real companies using this}

Step 5: Weighted Score & Ranking

Calculate weighted total per option per layer. Present ranking:

### {Layer} Ranking

1. **{Option A}** — Score: {X}/100 — Confidence: {Y}%
2. **{Option B}** — Score: {X}/100 — Confidence: {Y}%
3. **{Option C}** — Score: {X}/100 — Confidence: {Y}%

**Recommendation:** {Option A}
**Why:** {2-3 sentence rationale linking to team skills, scale, and constraints}

Step 6: Generate Report

Write report to {plan-dir}/research/tech-stack-comparison.md with:

Executive summary (recommended full stack in 5 lines)
Technical requirements table (from Step 2)
Per-layer comparison matrices (from Step 4)
Per-layer rankings with recommendations (from Step 5)
Combined recommended stack diagram
Risk assessment for recommended stack
Alternative stack (second-best combo) for comparison
Unresolved questions

Report must be <=200 lines. Use tables over prose.

Step 7: User Validation Interview

MANDATORY IMPORTANT MUST ATTENTION present findings and ask 5-8 questions via a direct user question:

Required Questions

Per-layer recommendation confirmation — "For {layer}, I recommend {option}. Agree?"
- Options: Agree (Recommended) | Prefer {option B} | Need more research
Risk tolerance — "The recommended stack has {risk}. Acceptable?"
Team readiness — "Team needs to learn {X}. Training plan needed?"
Budget alignment — "Estimated infra cost: ${X}/month. Within budget?"
Timeline fit — "This stack enables MVP in {X} months. Acceptable?"

Optional Deep-Dive Questions (pick 2-3 based on context)

"Should we consider {emerging tech} for {layer}?"
"Any compliance requirements I haven't captured?"
"Preference for managed services vs self-hosted?"
"Monorepo or polyrepo for this team size?"

After user confirms, update report with final decisions and mark as status: confirmed.

Output

{plan-dir}/research/tech-stack-comparison.md    # Full comparison report
{plan-dir}/phase-02-tech-stack.md               # Final confirmed tech stack decisions

MANDATORY IMPORTANT MUST ATTENTION break work into small todo tasks using task tracking BEFORE starting. MANDATORY IMPORTANT MUST ATTENTION validate EVERY recommendation with user via a direct user question — never auto-decide. MANDATORY IMPORTANT MUST ATTENTION include confidence % and evidence citations for all claims. MANDATORY IMPORTANT MUST ATTENTION add a final review todo task to verify work quality.

Next Steps

MANDATORY IMPORTANT MUST ATTENTION — NO EXCEPTIONS after completing this skill, you MUST ATTENTION use a direct user question to present these options. Do NOT skip because the task seems "simple" or "obvious" — the user decides:

"$architecture-design (Recommended)" — Design solution architecture with chosen tech stack
"$plan" — If architecture already decided
"Skip, continue manually" — user decides

Council escalation (always-offer, second prompt)

After the existing ## Next Steps prompt above resolves, present a second, independent a direct user question call:

"Skip council — proceed with chosen stack (Recommended)" — Continue with the selected tech stack as-is.
"Escalate to $llm-council" — Run 11 sub-agent council. Best applied when 2+ stacks score within 15% on the comparison matrix or you have unfamiliar/strategic dependencies. Cheaper alternatives: $why-review, $plan-validate.

Prompt-Enhance Closing Anchors

IMPORTANT MUST ATTENTION follow declared step order for this skill; NEVER skip, reorder, or merge steps without explicit user approval
IMPORTANT MUST ATTENTION for every step/sub-skill call: set in_progress before execution, set completed after execution
IMPORTANT MUST ATTENTION every skipped step MUST include explicit reason; every completed step MUST include concise evidence
IMPORTANT MUST ATTENTION if Task tools unavailable, maintain an equivalent step-by-step plan tracker with synchronized statuses

Closing Reminders

IMPORTANT MUST ATTENTION Goal: deliver user-confirmed, per-layer tech stack — each choice backed by 3+ researched options, weighted 8-criteria scoring, cited evidence, confidence % — so team commits to a stack fit for scale, budget, skills, timeline, NOT familiarity.

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

Critical Thinking: MUST ATTENTION apply critical + sequential thinking; traced proof, confidence >80% to act, NEVER guess as fact.
AI Mistake Prevention: verify generated content against evidence, trace downstream references, verify all affected outputs, re-read after context loss, surface ambiguity.

IMPORTANT MUST ATTENTION research minimum 3 WebSearched options per stack layer (backend, frontend, database, messaging, infra, auth); every recommendation carries confidence % + cited evidence (URL, benchmark, case study) — NEVER recommend on familiarity alone — why: familiarity bias commits the team to the wrong stack that surfaces only at scale. IMPORTANT MUST ATTENTION gate on user via a direct user question at EVERY decision point — confirm derived requirements before research (Step 2), confirm each layer recommendation in the end interview (Step 7) — NEVER auto-decide — why: the team owns the stack, not the AI. MANDATORY IMPORTANT MUST ATTENTION break work into small todo tasks using task tracking BEFORE starting; mark one in_progress, completed immediately after evidence; add a final review todo.

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.

[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.

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.

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.

IMPORTANT MUST ATTENTION requirements come BEFORE research — load prior business/domain/PBI artifacts (Step 1), map business signals to technical requirements (Step 2), user-confirm them, THEN WebSearch (Step 3) — NEVER research before requirements are derived and confirmed — why: researching first picks tech then back-fits the problem, the reverse of architecture. IMPORTANT MUST ATTENTION score every layer with the weighted 8-criteria matrix (High=3x / Medium=2x / Low=1x), rank with confidence %, cap the {plan-dir}/research/tech-stack-comparison.md report at <=200 lines using tables over prose — why: an unscored or unbounded report hides the trade-off the decision turns on. IMPORTANT MUST ATTENTION only user-confirmed decisions get written to phase-02-tech-stack.md as status: confirmed — the end interview (5-8 a direct user question questions) is mandatory and never skipped even when the choice seems "obvious" — why: an unconfirmed stack is a guess the team will pay for. IMPORTANT MUST ATTENTION every claim, finding, and recommendation requires file:line/URL proof or traced evidence + confidence % (>80% act, 60-80% verify first, <60% DO NOT recommend) — NEVER present a guess as fact — why: a stack chosen on speculation fails silently until production. IMPORTANT MUST ATTENTION evaluate fit before copying a reference stack from another project — verify the new context shares the same scale, budget, team skills, compliance, and timeline constraints — why: the closest example rarely matches preconditions, and a mismatched copy compiles but fails the real requirements.

Anti-Rationalization:

| Evasion | Rebuttal | | ------------------------------------------------ | ------------------------------------------------------------------------------------------ | | "Stack is obvious — skip the research" | 3+ WebSearched options per layer with cited evidence anyway — familiarity is not evidence. | | "I already know this is the best framework" | Show the weighted 8-criteria score + confidence %. No matrix = no recommendation. | | "Skip the user interview, the choice is clear" | The end interview is MANDATORY — only status: confirmed decisions get written. | | "Just research the stack, requirements are fine" | Derive + user-confirm technical requirements FIRST (Steps 1-2), then research. | | "One source is enough for this layer" | Cite URL + benchmark + case study; a single anecdote is not benchmarked evidence. |

External Memory: For research/analysis work, write intermediate findings and final results to a report file in plans/reports/ — prevents context loss and serves as deliverable.

Evidence Gate: MANDATORY IMPORTANT MUST ATTENTION — every claim, finding, recommendation requires file:line/URL proof or traced evidence with confidence percentage (>80% to act, <80% must verify first).

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.

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.
ANALYZE: Choose the best option: execute directly, invoke a skill, activate a standard workflow, or compose a custom step combination.
AUTO-SELECT: Pick the best option yourself. Do not ask the user to choose between direct execution, skill, standard workflow, or custom workflow.
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.
CREATE TASKS: task tracking for ALL workflow/skill/custom steps before execution when the selected path has multiple steps.
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.

Keep reusable AI-SDD principles in .claude; put repository-specific paths, commands, owners, products, and formats in project config/reference docs.

Preserve cycle: spec -> plan -> tasks -> implement -> verify -> update spec/docs.

Trace every requirement or invariant through decision, task, TC/test, source evidence, and docs/spec update.

Treat code-to-spec extraction as reference-only until accepted by the canonical spec owner.

Any supported AI tool may plan, implement, review, or verify with synced context; using multiple tools is optional.

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

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:

Name the FAILURE MODE (reasoning/assumption failure), not symptom — "assumed API existed without reading source" not "used wrong enum value".
Generality test: does this failure mode apply to ≥3 contexts/codebases? If not, abstract one level up.
Write as a universal rule — strip project-specific names/paths/classes. Useful on any codebase.
Consolidate: multiple mistakes sharing one failure mode → ONE lesson.
Recurrence gate: "Would this recur in future session WITHOUT this reminder?" — No → skip $learn.
Auto-fix gate: "Could $code-review/$code-simplifier/$security-review/$lint catch this?" — Yes → improve review skill instead.
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.

Agent Skills: tech-stack-research

Install this agent skill to your local

Skill Files