aida

Plugin aida from marketplace camoa-skills in camoa/claude-skills. Commands read from the repo's manifest — today.

Skills (13)

• completion

  This skill should be used when a reviewed task is ready to close, for example "close this task", "finish the task", "mark the task done", "write the pull request body", or "complete the task". It reads the review verdict and offers one follow up task per open finding. It writes a pull request body f

• design

  This skill should be used when a task's criteria are grounded and it is time to decide how to build them, for example "design this task", "write work orders", "architect this feature", "plan the build", or "Phase 2". It writes one work order per unit of build, each naming the criteria it serves and

• implement

  This skill should be used when a task's design has closed cleanly and it is time to begin building, for example "start implementing this task", "begin the build", or "Phase 3". It freezes the criteria and the work orders into a snapshot, opens the ledger that tracks each order's progress, refuses to

• next

  This skill should be used when the user asks "what's next", "what should I work on", "continue", "resume", "pick up where I left off", or names a task directly to jump to it. It lists which tasks are open in the current project and where each stands, or loads the one named, and offers to start a tas

• playbooks

  This skill should be used when the user asks "which playbooks apply", "list the plays", "add a play", "capture this as a rule", "add this to the playbook", "show my playbook", or when completion offers a note as a play and the person says yes. It lists the plays from the person's file, the project's

• project

  This skill should be used when the user asks "which project", wants to "create a project", "start a new project", "switch project", "mark this project complete", "archive a project", "unregister a project", "install the task rule", "set where task worktrees go", "check this machine", or "uninstall A

• research

  This skill should be used when a task's scope contract is approved and its criteria need grounding before design starts, for example "research this task", "find prior art", "check for an existing library", "look for a guide", "check this assumption", or "Phase 1". It fans out one small search per su

• review

  This skill should be used when a task's implementation has finished and the whole task needs one pass against its contract and its code, for example "review this task", "run the review", "gate check", "check this task before completion", or "Phase 4". It runs sixteen checks over the frozen contract,

• scope

  This skill should be used when a task needs its scope contract written or changed, for example "define scope", "write acceptance criteria", "what does this task have to do", "add a non-goal", "add a criterion", "change the contract", or "scope this task". It runs a conversation that produces alignme

• surfaces

  This skill should be used when end to end or visual regression has to be set up for the current project, for example "set up e2e", "set up visual regression", "register a surface", "remove a surface", "take the first baselines", "add a page to the visual review", or when a stage's offer, at scope, d

• task

  This skill should be used when the user wants to "create a task", "start a new task", "split a task", "make this an epic", "mark a task in progress", "mark a task done", "complete a task", "run this task autonomously", "set a budget on this task", "raise the budget", "save what we decided", "bring t

• test-cleanup

  This skill should be used when a project's test suite has grown too large and a person wants it analysed and cleaned up, for example "clean up the tests", "our suite is too big", "find exaggerated tests", "which tests can we delete", or "shrink the test suite". It sorts every test against one rubric

• tool

  This skill should be used when a tool has to be installed or run in the AIDA project that owns this directory, for example PHPUnit, PHPStan, Playwright or that project's own test runner, or when a stage needs a tool before it can do its work. It follows that tool's recipe for this project's framewor

Subagents (16)

• acceptance-verifier

  Runs one finished task's real product against its frozen contract and records what it saw, one evidence row per criterion and per changed surface at each viewport. Dispatched by the review skill only, at the acceptance step. Did not build the task, writes no code and no test, and never decides a cri

• architecture-reviewer

  Reviews one finished task's whole diff against its frozen contract, over eight named lenses, in one pass. Dispatched by the review skill only. Never writes to the code under review, never runs a command, and never decides what happens to a finding.

• catalog-identifier

  Asks the guide catalog which guides and recipes cover a name, and returns the names that matched. Dispatched by the research, design, implement and tool skills, and by any stage whose recipe-point call could not find the navigator. Identifies only, and never opens a guide body.

• design-critic

  Reads one task's closed contract, its research records and its work orders from disk, under one lens the dispatch names, and writes one findings file. Dispatched by the design skill only, three times in parallel, before the design closes. Never edits a work order and never decides whether the design

• disposition-confirmer

  Confirms a reuse decision that nobody watched, by reading the written reasoning and the files it cites. Dispatched by the design skill on the unattended path only, once per work order. Its only write is that order's verdict file.

• distiller

  Reads one stage's close record from disk and says whether it stands alone without the conversation that produced it. Dispatched by the scope, research and design skills at their close step, and by the task skill's save. A light task's design close names all three stages in one dispatch. Never edits

• fixer

  Fixes the open findings of one review round, inside a fix scope. Dispatched by the implementation skill's review step only. Never changes a test and never fixes a finding not on its list.

• implementer

  Writes production code for one unit of work until its frozen tests pass. Dispatched by the implementation skill only. Never changes a test and never writes outside the files its unit owns.

• internal-searcher

  Searches this project's own code and configuration for prior art, or reads the interface of one or more reuse candidates. Dispatched by the research and design skills only. Reads the project, never the web.

• outward-searcher

  Runs one search outside this project and reports what the search found. Dispatched by the research, design and review skills only, one subject per dispatch.

• playbook-loader

  Turns each catalog playbook set a project subscribes to into plays, one per catalog entry, and writes them to one record in the task folder. Dispatched by the research skill only, before `playbooks load`. Never opens a guide body.

• reviewer

  Reviews one unit's diff against its contract, or verifies one fix round, depending on the dispatch. Dispatched by the implementation skill's review step only. Never writes to the code under review and never runs the tests.

• row-checker

  Confirms or rejects the trace-matrix rows for one order's criteria, against the test-authoring recipe, in both run modes. Dispatched by the implementation skill's checkpoint step. Never reads production source and never opens the implementation.

• split-advisor

  Reads one task's closed research from disk and recommends, once, whether the task stays flat or splits into children. Dispatched by the research skill only, after the coverage check passes. Never splits anything; a person decides. Its only write is one sidecar.

• test-author

  Writes the tests for one unit of work from its criteria, watches each one fail, and stops. Dispatched by the implementation skill only. Never reads production source and never writes production code.

• test-reader

  Reads one group of a project's test files, sorts every test into one category of the cleanup rubric, and writes one report with counts, causes and proposals. Dispatched by the test-cleanup skill only, once per group. Never edits a test, never runs one, and never decides what is deleted.

Hooks (9)

• SessionStart

  "${CLAUDE_PLUGIN_ROOT}"/hooks/session-start.sh

• PreCompact

  "${CLAUDE_PLUGIN_ROOT}"/hooks/pre-compact.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/deny-prior-source.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/deny-frozen-test-writes.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/deny-destructive-commands.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/deny-named-role-dispatch.sh

• PostToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/replies.sh

• SubagentStart

  "${CLAUDE_PLUGIN_ROOT}"/hooks/replies.sh

• SubagentStop

  "${CLAUDE_PLUGIN_ROOT}"/hooks/replies.sh

Similar plugins