deftai/directive

Best practices and guidelines for AI agents in software development

This repository also configures its own agents. See what directive tells them →

32Stars on the repository
50Mods indexed here, across every type
yesterdayLast push, which is what freshness is scored on
customA LICENSE file GitHub cannot name, so bodies are not copied

directive

01

deftai/directive

Skill Claude CodeCodex

! Cold-start check: deft runs from the npm-installed engine (npm i -g @deftai/directive). If neither deft nor directive will run on this machine, do not proceed with the instructions below -- recover first (#1933 Option 1, deprecate-by-disuse). ! To recover: install/update the global launcher, then repair wiring — npm…

not rated 32 yesterday C 0 tokens

deftai/directive

Skill Claude CodeCodex

Evaluate an article, paper, or post for lessons that could improve directive. Analyzes two axes: how concepts improve directive's own implementation, and how they improve the projects directive creates. Produces filtered suggestions, iterates with the user, and optionally creates GitHub issues on the directive repo.

not rated 32 yesterday A 68 tokens

deft-directive-cost

03

deftai/directive

Skill Claude CodeCodex

Pre-build cost & budget transparency phase. Use after the project SPECIFICATION is approved and before the build skill kicks off, to produce a plain-English COST-ESTIMATE.md the user can read, react to, and explicitly accept, rescope, no-build, or skip with a recorded reason. Audience is non-technical users; voice is…

not rated 32 yesterday A 83 tokens

deftai/directive

Skill Claude CodeCodex

Systematic, evidence-based root-cause investigation. Use when the user asks "why did X break / slow down / regress?", wants to debug a failure, run a forensic investigation, or find the root cause of a bug -- to enter a sustained investigation MODE with a claim ledger, mandatory falsification before any conclusion, a…

not rated 32 yesterday A 121 tokens

deftai/directive

Skill Claude CodeCodex

Convert phase/epic scope vBRIEFs into swarm-ready story vBRIEFs before swarm allocation. Use when the operator asks to decompose, split stories, or assess swarm readiness.

not rated 32 yesterday A 46 tokens

deftai/directive

Skill Claude CodeCodex

Batched session-end gap escalation for directive consumers. Collects friction/gap reports, drafts deduped framework-gap issues against deftai/directive, and files upstream only after explicit operator confirmation.

not rated 32 yesterday A 45 tokens

deftai/directive

Skill Claude CodeCodex

Explore a codebase for architectural improvement opportunities -- deepen shallow modules, generate competing interface designs in parallel via sub-agents, and file a refactor RFC as a GitHub Issue. Use when improving architecture, finding refactoring opportunities, or making a codebase more testable and AI-navigable.…

not rated 32 yesterday A 76 tokens

deftai/directive

Skill Claude CodeCodex

Break a SPECIFICATION.md, PRD, or plan into independently-grabbable GitHub Issues using tracer-bullet vertical slices. Use after spec generation, when the user wants to create implementation tickets, or when breaking work into parallel-assignable issues. Requires the GitHub CLI (gh).

not rated 32 yesterday A 67 tokens

deftai/directive

Skill Claude CodeCodex

Extract a DDD-style ubiquitous language glossary from the current conversation, flagging ambiguities and proposing canonical terms. Saves to UBIQUITOUSLANGUAGE.md. Use when defining domain terms, building a glossary, hardening terminology, or when "domain model" or "DDD" is mentioned.

not rated 32 yesterday A 67 tokens

deftai/directive

Skill Claude CodeCodex

Adversarial one-question-per-turn plan stress-testing before vBRIEF or plan artifacts. Use when the operator asks to run probe or stress-test a plan.

not rated 32 yesterday A 39 tokens

deftai/directive

Skill Claude CodeCodex

Cut a v0.X.Y release of the deft framework safely. Use when the user says "release", "cut release", "v0.X.Y", or "publish release" -- to walk an 8-phase workflow that pre-flights, runs an end-to-end rehearsal against a temp repo, lands a draft GitHub release (npm ships irrevocably at tag push), optionally QA's draft…

not rated 32 yesterday A 117 tokens

deftai/directive

Skill Claude CodeCodex

Withdrawn (#4070). Classify playbook is off. Work selection is plan-sequence:current then a read-only triage:queue listing. Replacement sieve is #4071. Triggers still match so packs:slice finds this stub.

not rated 32 yesterday A 58 tokens

deftai/directive

Skill Claude CodeCodex

Runs the Directive Tutorial inside the current session. Use after the person first saves their Name, or when they ask to start, skip, resume, or re-run the tutorial (Signal, Postcard, or Echo). Do NOT trigger on session:start ceremony alone, triage:welcome, or a real product build.

not rated 32 yesterday A 70 tokens

deftai/directive

Skill Claude CodeCodex

Create new deft skills with proper structure, RFC2119 notation, triggers, and progressive disclosure. Use when user wants to create, write, or build a new deft skill.

not rated 32 yesterday A 43 tokens

deftai/directive

Skill Claude CodeCodex

Create and verify dense xBRIEF artifacts (json|md|both) at an explicit --out path. On-demand pack templates for scope/playbook/mission/project styles. Use when compressing SoT into xBRIEF or running xbrief:create / xbrief:verify. Do NOT trigger on scope lifecycle (promote/activate/complete).

not rated 32 yesterday A 77 tokens

deftai/directive

Skill Claude CodeCodex

Thin design-critique router. Before every issue:ingest: arc or record design-critique: not warranted, because … (default spend N=1; panel N>=3 = Stop 2). Use when the operator asks for a design critique, design-critique, critique panel, mechanism-shaped triage, arc, or run an arc. Do NOT restart during already-planned…

not rated 32 yesterday A 109 tokens

deftai/directive

Skill Claude CodeCodex

Thin router for Stage A issue evaluation: isolated origin/master validity, parent WIP census, named gitignored sink. Use when the operator asks to evaluate issues, run issue-eval, or triage:evaluate. Do NOT trigger on ordinary implement, build, swarm, or design-critique dispatch.

not rated 32 yesterday A 70 tokens

deftai/directive

Skill Claude CodeCodex

Consented agent-driven product check-in (pulse + portrait) with enable and consent gates, minimized local ledger attach, and private GitHub sink submit. Defaults off.

not rated 32 yesterday A 41 tokens

deftai/directive

Skill Claude CodeCodex

Brownfield specification reconstruction + advise|shadow|enforce drift guard (#1589 / #5350). Draft-only synthesis from completed xBRIEFs + #1595 codebase MAP; never auto-promotes. Use when reconstructing a failed/stale spec, recovering requirements, or checking semantic spec drift. Do NOT trigger on ordinary…

not rated 32 yesterday A 91 tokens

forensic-research

20

deftai/directive

Skill Claude CodeCodex

Exhaustive evidence-based investigation with sub-agents, vBRIEF claim ledger, and mandatory falsification before conclusions or recommendations. Domain-agnostic core with pluggable domain packs. Artifacts live under .tmp/investigations/ only. Use for "why did", "investigate", "root cause", "what caused", "forensic…

not rated 32 yesterday A 187 tokens

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: