Agent Claude Code
Review a just-completed scenario or task step with fresh, deliberately-unnarrowed eyes and surface any problem the work contains.
50 5d ago A 29 tokens
original MIT
Agent Claude Code
Merge a story's drafted scenarios that share one execution into single scenarios, preserving every assertion, before the set is tiered.
50 5d ago A 29 tokens
original MIT
Agent Claude Code
Analyze coverage gaps after GREEN phase, update progress.md with targeted red/green steps.
50 5d ago A 19 tokens
original MIT
Agent Claude Code
Review frontend components for hardcoded mockup placeholder data.
50 5d ago A 14 tokens
original MIT
green-agent
05Agent Claude Code
TDD Green Phase - Implement minimal code (tests are READ-ONLY).
50 5d ago A 18 tokens
original MIT
Agent Claude Code
Scan one artifact against exactly one hazard-catalogue group and report which fired triggers lack their forced guard.
50 5d ago A 26 tokens
original MIT
Agent Claude Code
Reconcile the seams between hazard-catalogue group passes and name the single guard that closes each one.
50 5d ago A 26 tokens
original MIT
Agent Claude Code
Imagine the incident a just-completed scenario or task step would cause, working backward to find the missing guard.
50 5d ago A 26 tokens
original MIT
Agent Claude Code
Scan prompt documentation files for layer violations and structural problems, then fix them.
50 5d ago A 19 tokens
original MIT
red-agent
10Agent Claude Code
TDD Red Phase - Write tests with predicted failure.
50 5d ago A 13 tokens
original MIT
Agent Claude Code
Gather refactor detector findings and apply refactorings serially, one at a time.
50 5d ago A 21 tokens
original MIT
Agent Claude Code
Detect domain-modeling / behavior-placement smells (read-only) — cluster D of the refactor scan.
50 5d ago A 25 tokens
original MIT
Agent Claude Code
Detect duplication, test-DSL and frontend smells (read-only) — cluster T of the refactor scan.
50 5d ago A 28 tokens
original MIT
Agent Claude Code
Detect structural-mechanics smells (read-only) — cluster M of the refactor scan.
50 5d ago A 23 tokens
original MIT
Agent Claude Code
Gather detector findings and apply strict-assertion fixes serially.
50 5d ago A 16 tokens
original MIT
Agent Claude Code
Detect loose/weak test assertions (read-only) — cluster A of the test-review checklist.
50 5d ago A 25 tokens
original MIT
Agent Claude Code
Detect test-class vs Statements placement / DSL-separation violations (read-only) — cluster P.
50 5d ago A 23 tokens
original MIT
Agent Claude Code
Detect Selenium navigation / spec-depth / display-annotation violations (read-only) — cluster Se.
50 5d ago A 24 tokens
original MIT
Agent Claude Code
Detect Statements-internal quality violations (read-only) — cluster S.
50 5d ago A 19 tokens
original MIT
test-runner
20Agent Claude Code
Execute tests for specified module.
50 5d ago A 9 tokens
original MIT
Agent Claude Code
Sort a story's drafted scenarios into delivery tiers by consequence of failure, write their Tier: markers, and move Tier 3 out of the plan.
50 5d ago A 33 tokens
original MIT
PreToolUse
22Hook Claude Code
Runs before the agent uses a tool for Agent tool calls, executing agent-timing.sh via bash. From rakovi4/continue-framework.
50 5d ago A
tokens not measured
original MIT
PostToolUse
23Hook Claude Code
Runs after a tool call finishes for Agent tool calls, executing agent-timing.sh via bash. From rakovi4/continue-framework.
50 5d ago A
tokens not measured
copy · 89% MIT
Hook Claude Code
Runs when a tool call fails for Agent tool calls, executing agent-timing.sh via bash. From rakovi4/continue-framework.
50 5d ago A
tokens not measured
original MIT