Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/theafh/ai-modules/task_auto_checknpx skills add theafh/ai-modules --skill task_auto_checkgit clone --depth 1 https://github.com/theafh/ai-modulesWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/theafh/ai-modules/task_auto_check)<a href="https://agentmods.dev/skills/theafh/ai-modules/task_auto_check"><img src="https://agentmods.dev/badge/skills/theafh/ai-modules/task_auto_check.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00091 | $0.06517 |
| Opus 5 | $0.00046 | $0.03259 |
| Sonnet 5 | $0.00018 | $0.01303 |
| Haiku 4.5 | $0.00009 | $0.00652 |
Grade A, and why
task_auto_check scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 184 lines — stays where its author put it; the contents beside it link to each section on GitHub.
task_auto_check
<task_auto_check_skill>
<when_to_activate> Activate when the user points at one task and asks for the task itself to be made implementation-ready:
- "Auto-check this task until it is ready."
- "Make
<task>ready without implementing it." - "Run the autonomous readiness loop on
<task>." - "Fix the readiness issues from task_check if they are safe."
Route to task_check when the user wants a single read-only readiness verdict. Route to the base task skill or task_create when the user wants to write or manually edit a task. Route to task_implement when the user wants the task's described code/docs work built. Route to task_finish for close-out and archive moves.
</when_to_activate>
<path_resolution>
Resolve the base task and task_check skills from the same plugin bundle as this skill when possible: this skill lives at skills/task_auto_check/SKILL.md, so sibling skills live under ../task/ and ../task_check/. Resolve the helper agents by their published names — auto_drift_task, auto_gate_task, auto_reviewer_task, and auto_verifier_task — using the current harness's normal agent mechanism. When a harness exposes only file paths, those agents live in the same plugin at ../../agents/ relative to this skill directory.
</path_resolution>
- A max-round override, expressed as a positive integer such as "max rounds 2".
- Creation-time intent context, used only when the loop runs immediately after a task is drafted.
- An explicit request to use an available foreign-model reviewer stance. The default is single-model operation with no foreign-model stance.
<loop_policy>
<single_gate>
Use task_check verbatim as the gate. The loop consumes the structured verdict returned by auto_gate_task: task path, status stamp, prior status, ready boolean, per-item checklist record, issue list, and evidence labels. It does not compute a second readiness score and does not override task_check's ready or checked stamp.
</single_gate>
<frozen_intent>
Freeze the original task's # Title and ## Goal before the first gate call. When the user supplied creation-time intent, freeze that prompt alongside the title and Goal. Every proposal and every applied edit must preserve this frozen intent; the loop may clarify expression, add missing implementation context, or make acceptance checks verifiable, but it must not change what the task is for.
</frozen_intent>
<intent_drift_boundary>
Invoke auto_drift_task once at freeze time, before the first <gate> call and any repair. A drift classification is human-routed through the same surfaced stuck channel as <structural_split_boundary> and <mechanical_lint_boundary>: report the intention check that names the field auto_drift_task flagged in its drifted_fields — Attention: this task's Title appears to have already drifted from its original intent. for title-only drift, …this task's Goal appears… for goal-only drift, or …this task's Title and Goal appear… when both drifted — with the recovered-versus-current evidence, halt the auto-repair path for this run, leave the task body unchanged, and keep <frozen_intent> intact. Use the recovered origin as evidence for the human, not as an edit target; the loop never auto-repairs toward the recovered original intent. Clean, meaning-preserving, and low_confidence_clean results proceed without surfacing the intention check.
</intent_drift_boundary>
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 184 lines · 91 tokens per session scan A 59986ed68745
task_auto_check is a skill published in the GitHub repository theafh/ai-modules (38 stars, last pushed 4d ago), licensed MIT. It adds 91 tokens to every session and 6,517 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
wayfinder
Plan a huge chunk of work (more than one agent session can hold) as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.
sap-transport-overview
System-wide inventory of open transport requests — every visible modifiable request, who owns it, how big it is, and supported risk signals such as empty/local requests, explicit locks, and confirmed manifest overlaps. Headers-only and cheap; NO source diffs. Use when asked "what transports are open in the system"…
timeline-creator
Create HTML timelines and project roadmaps with Gantt charts, milestones, phase groupings, and progress indicators. Use when users request timelines, roadmaps, Gantt charts, project schedules, or milestone visualizations.
pm-all
Skill "pm-all" from wei63w/pm-manager, covering user input, outline and shared workflow (all /pm- commands).
bootstrap-prd
Set up PRD-driven development infrastructure for a new project, including directory structure, templates, and roadmap.
finn-spec
Interview the user about a raw idea until confident, then file a build-ready issue in Linear. Use when asked to run Finn-loop's spec interview, draft a queue-ready issue, or plan a feature. Interactive — requires the user present; never run unattended.