crewrig

Pull request and logbook composer. Activate when opening a PR, updating a PR description, or appending to a logbook issue. Produces titles, bodies, test plans, and logbook entries that conform to the project's AGENTS.md conventions.

Pull request and logbook composer. Activate when opening a PR, updating a PR description, or appending to a logbook issue. Produces titles, bodies, test plans, and logbook entries that conform to the project's AGENTS.md conventions.

Pull request and logbook composer. Activate when opening a PR, updating a PR description, or appending to a logbook issue. Produces titles, bodies, test plans, and logbook entries that conform to the project's AGENTS.md conventions.

Implementation skill for writing, modifying, and refactoring code. Activate by default for any coding task that does not warrant the architect skill. Optimised for parallelisable execution, fast feedback loops, and minimal surface area per change.

Implementation skill for writing, modifying, and refactoring code. Activate by default for any coding task that does not warrant the architect skill. Optimised for parallelisable execution, fast feedback loops, and minimal surface area per change.

Implementation skill for writing, modifying, and refactoring code. Activate by default for any coding task that does not warrant the architect skill. Optimised for parallelisable execution, fast feedback loops, and minimal surface area per change.

Independent PR review skill. Activate to audit a pull request cold — without authoring context — covering correctness, convention compliance, test coverage, and linter findings. Emits a structured verdict (Approve / Request Changes / Comment).

Independent PR review skill. Activate to audit a pull request cold — without authoring context — covering correctness, convention compliance, test coverage, and linter findings. Emits a structured verdict (Approve / Request Changes / Comment).

Independent PR review skill. Activate to audit a pull request cold — without authoring context — covering correctness, convention compliance, test coverage, and linter findings. Emits a structured verdict (Approve / Request Changes / Comment).

Harness feedback-loop curator. Activate on demand to read friction tags from the global harness-friction wing, cluster them, and open one descriptive issue per cluster on the canonical/feedback repos declared in components' provenance blocks. The fix MR lands later (human-authored or via the auto-fix mode tracked in #42). Also supports a `--deep` mode that sweeps `wing=transcripts` with heuristic pre-filtering and emits a Markdown review document for triage. Auto mode (#42) supports scheduled runs via scripts/schedule-curator.sh with dedup and per-run issue cap.

Tag a friction encountered during real work. Activate the moment any recognition signal fires (user pushback, second-time tool surprise, sibling-skill workaround, etc.). The single canonical implementation of the friction-tagging protocol — all other skills point here.

Harness feedback-loop curator. Activate on demand to read friction tags from the global harness-friction wing, cluster them, and open one descriptive issue per cluster on the canonical/feedback repos declared in components' provenance blocks. The fix MR lands later (human-authored or via the auto-fix mode tracked in #42). Also supports a `--deep` mode that sweeps `wing=transcripts` with heuristic pre-filtering and emits a Markdown review document for triage. Auto mode (#42) supports scheduled runs via scripts/schedule-curator.sh with dedup and per-run issue cap.

Tag a friction encountered during real work. Activate the moment any recognition signal fires (user pushback, second-time tool surprise, sibling-skill workaround, etc.). The single canonical implementation of the friction-tagging protocol — all other skills point here.

Harness feedback-loop curator. Activate on demand to read friction tags from the global harness-friction wing, cluster them, and open one descriptive issue per cluster on the canonical/feedback repos declared in components' provenance blocks. The fix MR lands later (human-authored or via the auto-fix mode tracked in #42). Also supports a `--deep` mode that sweeps `wing=transcripts` with heuristic pre-filtering and emits a Markdown review document for triage. Auto mode (#42) supports scheduled runs via scripts/schedule-curator.sh with dedup and per-run issue cap.

Tag a friction encountered during real work. Activate the moment any recognition signal fires (user pushback, second-time tool surprise, sibling-skill workaround, etc.). The single canonical implementation of the friction-tagging protocol — all other skills point here.

Design and architecture skill for ADRs, RFCs, design reviews, and ripple-effect analysis. Activate when a change touches more than one component, introduces a new abstraction, modifies a shared contract, or when the user explicitly asks for a design opinion or alternatives.

Structured approach for building, debugging, and hardening CI/CD pipelines. Activate when the agent needs to: create new workflow definitions, diagnose failing jobs, validate pipeline changes on a branch before merge, or integrate platform-specific resources (secrets, registries, certificates).

Documentation skill for ADRs, READMEs, in-code docstrings, and reference material. Activate when the user asks for documentation, when a public contract changes without docs, or when an ADR is needed per the architect skill's output. Optimised for documents that age well.

Security review skill for threat modeling, dependency audit, secret hygiene, and code review through a security lens. Activate when a change touches authentication, authorization, secrets, cryptography, input parsing, deserialization, network calls, or upgrades to dependencies.

Test authoring and test-strategy skill. Activate when writing new tests, planning a test strategy, enumerating edge cases for a feature, or reviewing whether a change has adequate coverage. Optimised for high-signal tests that catch regressions, not coverage theatre.

Design and architecture skill for ADRs, RFCs, design reviews, and ripple-effect analysis. Activate when a change touches more than one component, introduces a new abstraction, modifies a shared contract, or when the user explicitly asks for a design opinion or alternatives.

Structured approach for building, debugging, and hardening CI/CD pipelines. Activate when the agent needs to: create new workflow definitions, diagnose failing jobs, validate pipeline changes on a branch before merge, or integrate platform-specific resources (secrets, registries, certificates).

Documentation skill for ADRs, READMEs, in-code docstrings, and reference material. Activate when the user asks for documentation, when a public contract changes without docs, or when an ADR is needed per the architect skill's output. Optimised for documents that age well.

Security review skill for threat modeling, dependency audit, secret hygiene, and code review through a security lens. Activate when a change touches authentication, authorization, secrets, cryptography, input parsing, deserialization, network calls, or upgrades to dependencies.

Test authoring and test-strategy skill. Activate when writing new tests, planning a test strategy, enumerating edge cases for a feature, or reviewing whether a change has adequate coverage. Optimised for high-signal tests that catch regressions, not coverage theatre.

Design and architecture skill for ADRs, RFCs, design reviews, and ripple-effect analysis. Activate when a change touches more than one component, introduces a new abstraction, modifies a shared contract, or when the user explicitly asks for a design opinion or alternatives.

Structured approach for building, debugging, and hardening CI/CD pipelines. Activate when the agent needs to: create new workflow definitions, diagnose failing jobs, validate pipeline changes on a branch before merge, or integrate platform-specific resources (secrets, registries, certificates).

Documentation skill for ADRs, READMEs, in-code docstrings, and reference material. Activate when the user asks for documentation, when a public contract changes without docs, or when an ADR is needed per the architect skill's output. Optimised for documents that age well.

Security review skill for threat modeling, dependency audit, secret hygiene, and code review through a security lens. Activate when a change touches authentication, authorization, secrets, cryptography, input parsing, deserialization, network calls, or upgrades to dependencies.

Test authoring and test-strategy skill. Activate when writing new tests, planning a test strategy, enumerating edge cases for a feature, or reviewing whether a change has adequate coverage. Optimised for high-signal tests that catch regressions, not coverage theatre.

Helps users introduce themselves using the hello-world MCP tool.