ai_dev

Plugin ai_dev from marketplace ai-modules in theafh/ai-modules. Commands read from the repo's manifest — today.

Skills (26)

• agent_spinner

  Run helper agents from an orchestrating agent under one quality contract, whatever the host offers. Establishes the capability tier, declares a phase plan, picks an orchestration shape and width, contains each helper to one path, checks work refute-by-default, and merges cited findings into one verd

• ai_instruction_formatting

  Organize AI-consumed content (prompts, rules, skills, commands, agents, system instructions) into pseudo-XML by wrapping each semantic concern in a dedicated tag for role, policy, inputs, and output contract. Use when structuring or reorganizing the body of a SKILL.md, agent or sub-agent definition,

• ai_instruction_writing

  Write AI-consumed content using positive, action-oriented language as the primary carrier of every instruction. Use when creating or editing any artifact an AI reads at inference time. Those artifacts include SKILL.md files, .mdc rule files, CLAUDE.md/AGENTS.md/GEMINI.md configuration, prompt templa

• format_markdown

  Apply markdown linting compliance and best practices when creating or editing markdown files (.md, .mdc). Covers blank-line rules around block elements, consistent bullet style, fenced code blocks with language identifiers, table alignment syntax, header level progression, list indentation and spaci

• format_python

  Apply formatting standards, code quality rules, structure conventions, linting prevention, and best practices when generating or editing Python code (.py). Covers indentation, quoting, imports, naming, line length, type hints, error handling, logging, data structures, testing, and security practices

• format_rust

  Apply clippy-aligned Rust practices when writing or editing Rust code (.rs). Covers procedural flow, clippy-driven clarity improvements, minimal imports, Result/Option idioms, the error-versus-invariant model with panic discipline and clippy unwrap_used enforcement, fallible builders following the e

• git_auto_review

  Run an automated review-and-fix loop over the checked-out branch's change since the default branch plus its working tree, and over the default branch's uncommitted changes: git_review is the only gate, a single writer agent fixes the findings that qualify, each round is verified refute-by-default an

• git_checkout

  Put the repository onto an existing branch wherever it lives: fetch every remote without pruning, switch when the branch is already local, create a local branch with an explicit upstream when one remote carries the name, ask which remote to track when several do, and separate a nonexistent branch fr

• git_commit

  Create one structured git commit covering the full current working-tree state. Use when the user explicitly asks to commit in the current turn, saying "commit", "commit this", "commit changes", "make a commit", "write a commit message", "git commit", or an equally direct ask to put the current repo

• git_refresh

  Refresh a local git repo back to a clean current default branch: fetch/prune, detect the remote HEAD, fast-forward main/master by --ff-only, delete merged local branches, and offer gated cleanup for upstream-gone, squash-merged, stale, or force-delete branch pruning. Use when the user asks to clean

• git_review

  Review the changes the user names and answer two questions kept apart: could they be approved in general, and can they be structurally merged. The target is a pull request by URL or number, a branch by name or checked out, or the default branch's uncommitted changes. The report opens with the review

• guardrail

  Hub and source of truth for the repo-root guardrail documents (CHARTER.md, ARCHITECTURE.md, TESTING.md, SECURITY.md) that keep AI agents anchored to human intent while a project evolves within its declared boundaries. Explains the doc set, the authority hierarchy, the shared format, and how skills c

• guardrail_audit

  Read-only audit of a repository's existing guardrail docs. It surfaces doc-vs-doc contradictions and doc-vs-code divergences ranked by the guardrail hierarchy, plus grounded missing-doc proposals where repo substance warrants an absent doc. Edits nothing; every finding carries evidence and a reconci

• harness_portability

  Cross-agent-harness and cross-OS portability review for runtime artefacts bundled inside skills and plugins. Use when creating, editing, reviewing, or troubleshooting shell scripts, Python or Node helpers, hooks and hook configuration, MCP servers, command wrappers, setup and install flows, plugin w

• skill_doctor

  Check-only doctor for skill artifacts. It audits SKILL.md frontmatter, descriptions that must serve both a browsing user and an LLM router, registration, tests, and instruction quality for one skill, a skill family, or every skill in the repo, editing no targets and running no harness-portability re

• task

  Manage the project task backlog as plain markdown files in tasks/. Use when the user asks to list or query existing tasks, to update or expand one, to triage or split backlog items, or to work across several tasks at once, and as the fallback for any tasks, todos, or backlog request naming no single

• task_audit

  Audit one implemented or finished task against the actual codebase. Use when verifying a believed-done task, checking claimed completion against the code, or running a drift check before close-out. Inspect code and tests, run verification, stamp audited only on clean implemented work, and report gap

• task_auto_check

  Autonomously drive one task from open or checked to ready through task_check, verifier-approved body repairs, and final mechanical task-lint cleanup. Use when a user asks to auto-fix readiness issues, make a task ready, refresh a stale task against the current codebase, or run an autonomous readines

• task_check

  Check one task before implementation. Use when the user asks if a task is ready to build or still valid. Assess structure, scope, focus, complexity, contradictions, and ambiguity, verify the task's premise and described current state against the codebase, surface a stale invalidated task with deferr

• task_create

  Create exactly one new task file for one upcoming work item. Use when the user asks to make, add, file, write, or capture a single task or todo. Use task for listing, querying, updating, archiving, linting, or multi task backlog work.

• task_explain

  Explain one project task in compact what/why/how prose without editing it. Use when the user asks to explain, summarize, walk through, or orient on a task file by path, partial name, or conversation context. If the task reference is missing or ambiguous, activate and ask which task to explain. Not f

• task_finish

  Close one completed or parked task. Use when the user asks to finish, mark done, defer, park, drop, or archive a task. Set finished or deferred, move the file to archive, update links, and relint.

• task_fix

  Repair the whole tasks backlog tree in one pass and assess its backlog-coherence by default. Use when the user asks to health check, clean up, audit, or lint the backlog, to check backlog coherence or consistency, to say whether selected tasks ship together, or to autonomously resolve backlog judgem

• task_implement

  Implement one existing task file end to end. Use when the user asks to build, do, or implement the work described by a task. Edit code and tests, run verification, stamp implemented, and leave audit and finish to sibling skills.

• task_select

  Select and rank eligible live tasks from the project backlog. Use when the user asks what task to work on next, asks the agent to pick or prioritize backlog work, rank open tasks, choose from tasks/, or recommend the next task/action without editing task files.

• update_changelog

  Create or update a day-grouped, add-only CHANGELOG.md from git commit history. Use when asked to generate, refresh, or maintain a repository changelog. Produces newest-first day sections while keeping existing entries immutable and processes one day at a time to stay within context limits.

Subagents (11)

• auto_checker_spinner

  Read-only checker role for an orchestrating agent; the orchestrator spawns it through a rendered brief, and it reports against that brief's own contract.

• auto_drift_task

  Reconstructs one task's last-commit title/Goal/body baseline for task_auto_check and reports meaning-level Goal/title drift only when the working tree broadens what the task is for; read-only and human-routed.

• auto_dryrun_task

  Take the implementer's seat on one task for task_auto_check before the gate of every ordinary repair round. Run the dry run by writing down the steps implementing the task exactly as written would take, walk those steps and the whole task (or, from the second round, only the open blocking issues and

• auto_fixer_git

  The single serialized writer for git_auto_review. For each approved finding it makes the fix in the working tree at that finding's location and adds a reproduction file directly in the run's ignored .git_auto_review folder at the path its ledger line names, or extends the reproduction that line alre

• auto_gate_git

  Runs git_review as the review gate for git_auto_review, keeping that skill's review criteria, headings, labels, and tags verbatim while forbidding its mutation paths, then derives a structured verdict from the report. Each block carries one finding's id, heading, label, non_blocking, location, lane,

• auto_gate_task

  Wraps task_check for task_auto_check by applying the single readiness gate to one task and returning task_check's full report followed by a structured verdict derived from it. The verdict carries final status, prior status, ready boolean, scope, blocking_count, per-item checklist record, issue list,

• auto_reviewer_task

  Proposes minimum task-body repairs for task_auto_check and read-side task_fix escalation, citing the base task skill's body repair rules and preserving frozen task intent.

• auto_shaper_task

  Resolves task_fix's escalated whole-tree judgement calls as the single serialized writer. It verifies splits, body-framing reframes, scope relocations, link repairs, repeated-link regroups, and the backlog-coherence repairs task_fix's default assessment surfaced and the user accepted, all against fr

• auto_verifier_git

  Refute-by-default verifier for git_auto_review. In triage mode it judges each qualifying finding of the round's gate verdict and approves one only when the fix its Fix names is real, minimum, scope-anchored, and safe, returning not-approved with a reason otherwise; in delta mode it maps every hunk o

• auto_verifier_task

  Refute-by-default verifier for task_auto_check and auto_shaper_task proposals, task_auto_check citation sets, and task_auto_check applied deltas. Keeps only real, minimum, issue-resolving, frozen-intent-preserving task edits or writer-executed structural plans; in citation mode keeps only clean clai

• auto_writer_spinner

  Writer role for an orchestrating agent; the orchestrator spawns it through a rendered brief, and it carries out that brief's one item.

Hooks (3)

• PreToolUse

  ${CLAUDE_PLUGIN_ROOT}/hooks/charter_guardrail.sh

• PreToolUse

  ${CLAUDE_PLUGIN_ROOT}/hooks/charter_guardrail.sh

• PreToolUse

  ${CLAUDE_PLUGIN_ROOT}/hooks/charter_guardrail.sh

Similar plugins