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 skills add blakecyze/kanso --skill kanso-taskgit clone --depth 1 https://github.com/blakecyze/kansoWrote 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/blakecyze/kanso/kanso-task)<a href="https://agentmods.dev/skills/blakecyze/kanso/kanso-task"><img src="https://agentmods.dev/badge/skills/blakecyze/kanso/kanso-task.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 5.1 | $0.00045 | $0.02669 |
| Opus 5 | $0.00023 | $0.01334 |
| Sonnet 5 | $0.00009 | $0.00534 |
| Haiku 4.5 | $0.00005 | $0.00267 |
Grade A, and why
kanso-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 7d 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 — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
kanso-task
The user types /kanso-task <rough request>. This skill turns that rough request into a sharp prompt, then executes the task in the same chat with both kanso-prompting and kanso-principles standing as context. The output should be measurably better than what the user would have got from typing the same words into the chat directly.
Three phases:
- Phase A — Clarify (only if needed). Run the ambiguity pre-flight. If a trigger fires, ask one tight batch of questions and fold the answers in. If the input is already concrete — or the user waved the skip valve — go straight to formatting with no questions.
- Phase B — Rewrite. Apply the
kanso-promptingrules to produce a sharp version of the prompt. Show it to the user. Wait for approval. - Phase C — Execute. Run the rewritten prompt inline in the current chat. Any code produced during execution follows
kanso-principles.
Always run inline
This skill runs in the calling chat. Never dispatch the rewrite, the approval gate, or the execution via the Agent or Task tool. Never spawn a subagent. The user has to be able to see the rewritten prompt before it runs, intervene during execution, and inspect the result without switching windows. A subagent dispatch defeats the entire workflow.
Phase A — Clarify
A pre-flight ambiguity check that runs before formatting. Its only job is to catch the gaps that would force a guess. Concrete input passes through untouched — the yap-and-go feel survives; the pause is earned, not routine.
The skip valve — check this first
Two ways an input skips questioning entirely:
- Explicit. The user signalled skip. The unambiguous form is the
--goflag. A plain-language equivalent (just format it,no questions) counts only when it is a directive about the request rather than part of it — a trailing or standalone instruction to the skill, not a phrase inside the task being described./kanso-task format the users table, no questions about the schema yetis a task that happens to contain "no questions"; it is not a skip signal. When a plain-language form is genuinely ambiguous, treat it as task content and run the triggers. Honour a real skip without comment: format silently, even if a trigger would otherwise fire. The user has taken the wheel. - Inferred. The input is high-confidence — a trained reader could produce a sharp prompt from it without inventing anything load-bearing. Format silently.
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.
- 7d ago First seen · 159 lines · 45 tokens per session scan A 27cde1504709
kanso-task is a skill published in the GitHub repository blakecyze/kanso (3 stars, last pushed 4d ago), licensed MIT. It adds 45 tokens to every session and 2,669 once invoked, about $0.0002 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
auto-loop
TDD-based autonomous development loop with checkpoint recovery and observability changelog.
hook-template
Generate hook script from template. Use when adding a new hook, wiring a PreToolUse/PostToolUse/Stop/Notification hook, or scaffolding hook config for settings.json.
agent-check
Validate custom agent file format and structure. Use after creating or editing an agent, before committing agent changes, or when an agent fails to load.
session-relay
Continue real repository work between Claude Code, Codex CLI, and Grok Build with a privacy-bounded, drift-aware handoff packet. Use for cross-vendor takeover, context-limit checkpoints, multi-hop Claude→Codex→Grok relays, or receiving an existing Director packet. Use native resume commands for same-CLI history; never…
skill-check
Validate skill/command file format and structure. Use after creating or editing a skill, before committing skill changes, or when a skill fails to load or trigger.
skill-template
Generate custom skill/command from template. Use when creating a new skill or slash command from scratch, or scaffolding a skill file with correct frontmatter.