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/ssbun/csl-agent-kit/tasknpx skills add SSBun/csl-agent-kit --skill taskgit clone --depth 1 https://github.com/SSBun/csl-agent-kitWhat 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 | $0.00079 | $0.01851 |
| Opus 5 | $0.00039 | $0.00925 |
| Sonnet 5 | $0.00016 | $0.00370 |
| Haiku 4.5 | $0.00008 | $0.00185 |
Grade A, and why
task 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 2d 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 — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Task
Manage one outcome through the host Agent. Use the host's existing file, shell, research, and subagent capabilities; never launch nested codex exec, pi --print, a worker daemon, or an unattended supervisor.
Required Runtime Protocol
Resolve the collection root as the parent of this skill directory. Before forming, presenting, or accepting a Task Target, read <collection-root>/csl-tasks/shared/protocols/task-target-alignment.md in full and treat it as the authoritative detailed alignment contract. Read it again after resume or compaction when it is no longer present in context. If it is unavailable, stop before substantive work and report the missing runtime dependency.
This skill owns task activation, the single-outcome Target meaning, permitted lifecycle writes, and the post-alignment task workflow. The shared protocol owns all Task Target alignment semantics.
Storage and Core
tasks/tasks.mdis the newest-first index. It contains only title, current status, and atasks/<slug>.mdlink.tasks/tasks/<slug>.mdis the canonical record and is authoritative when the index disagrees.- Use
node <collection-root>/csl-tasks/shared/scripts/csl-tasks.js --workspace <workspace> ...for creation, state, parent-child, evidence, verification, completion, and index checks. - Do not hand-edit status or index lines. After direct edits to Scope, Target, Plan, Decisions, or Block, run
sync <id>andcheck <id>. - Modify only the owning task and its exact index entry. Preserve unrelated task state and untouched history.
Activation and Ownership
- Create, resume, or reopen the owning record as soon as the request either asks to create, modify, move, rename, or delete any file, or establishes a concrete, non-trivial, independently acceptable outcome.
- Activate it before substantive discussion, repository exploration, research, planning, delegation, implementation, or any focused target-forming clarification performed after the outcome exists. For a requested file mutation, activation and—when the host provides one—an actually emitted
task_focus(<id>)call must also precede the first requested deliverable edit. Before activation, allow only Project Core loading, the minimal index and candidate-record lookup needed to resolve ownership, and the one focused question permitted by the shared protocol when no honest Target yet exists. - The lifecycle writes needed to create, resume, reopen, focus, sync, check, and align the owning task are the bootstrap exception to the first-edit rule; they do not authorize editing the requested deliverable.
- Start a new task for every independently acceptable user outcome.
- Reopen an existing task only when the request directly corrects, completes, or re-verifies the same outcome and leaving its Target or Result unchanged would be misleading.
- Component, file, topic, or implementation overlap alone does not establish ownership. Create a new task when ownership is ambiguous.
- Skip records only for simple factual answers and open-ended conversation that request no file mutation and establish no concrete outcome, read-only trivial deterministic operations, and routine Context or Lessons maintenance that is not itself the requested deliverable.
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2d ago First seen · 98 lines · 79 tokens per session scan A caf8fdb49c62
task is a skill published in the GitHub repository SSBun/csl-agent-kit (10 stars, last pushed 3d ago), licensed MIT. It adds 79 tokens to every session and 1,851 once invoked, about $0.0004 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…