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 Windy3f3f3f3f/claude-self-goal --skill skillgit clone --depth 1 https://github.com/Windy3f3f3f3f/claude-self-goalWrote 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/windy3f3f3f3f/claude-self-goal/skill)<a href="https://agentmods.dev/skills/windy3f3f3f3f/claude-self-goal/skill"><img src="https://agentmods.dev/badge/skills/windy3f3f3f3f/claude-self-goal/skill/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/windy3f3f3f3f/claude-self-goal/skill"><img src="https://agentmods.dev/badge/skills/windy3f3f3f3f/claude-self-goal/skill.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00092 | $0.00527 |
| Opus 5 | $0.00046 | $0.00264 |
| Sonnet 5 | $0.00018 | $0.00105 |
| Haiku 4.5 | $0.00009 | $0.00053 |
Grade A, and why
self-goal 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 10d 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.
What it actually says
self-goal
Give the current session its own native /goal by running the claude-self-goal tool from your Bash tool. It figures out the session type (interactive / tmux / daemon background) and picks the right delivery channel automatically.
When to use
Use this when, mid-task, you conclude you should keep going until a specific, checkable condition holds — and you want Claude Code's built-in goal evaluator (not just your own intention) to keep the session from stopping early.
Good conditions are objective: "all tests in ./test pass", "the file dist/app.js exists", "the build is green". Avoid vague ones like "the code is good".
How
# set a goal on this session (always quote the condition):
claude-self-goal "all tests pass and the build is green"
# clear it early if the plan changes:
claude-self-goal --clear
# set a goal on another background session you own:
claude-self-goal --session <id> "the PR's CI is green"
Replace claude-self-goal with the absolute path if it is not on PATH (e.g. /usr/local/bin/claude-self-goal).
Notes
- Auto-detected channel: tmux →
tmux send-keys(no root); plain interactive →TIOCSTI(needs root); daemon/bgsession →claude attachover its rv socket. A headlessclaude -psession can't take an interactive/goaland is refused. - Always quote the condition — an unquoted multi-word condition is split into several arguments by the shell.
- The condition is injected as a single line; control characters are rejected.
- Once set, the
◎ /goal activechip appears and Claude Code's native goal evaluator should keep the session working until the condition is evaluated as met. It auto-clears on success — do not tell the user to run/goal clearafter the goal is achieved.
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.
- 10d ago First seen · 37 lines · 92 tokens per session scan A 006b3412c509
self-goal is a skill published in the GitHub repository Windy3f3f3f3f/claude-self-goal (102 stars, last pushed 2mo ago), licensed MIT. It adds 92 tokens to every session and 527 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
orbit-notion
Open Orbit briefing skill — selected by the Orbit pipeline when Notion is the user's only connected connector, or when the user explicitly scopes their daily digest to Notion. Pulls the past 24 hours of document edits, comments, mentions, and database row changes from the user's authenticated Notion connection and…
Cortex
Operate Cortex, the LifeOS memory system — the typed Knowledge Archive (People, Companies, Ideas, Research with typed related: links) plus recall of prior work sessions, ISAs, and conversations. Search, add, harvest, develop, ingest, distill, graph-navigate, recall. USE WHEN cortex, knowledge, knowledge base, search…
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
peekaboo
Capture and automate macOS UI with the Peekaboo CLI.
mochi-remind
Handle due reminders — notify the user with natural language and mark them done.