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/msolecki/founder-os/scope-guardnpx skills add msolecki/founder-os --skill scope-guardgit clone --depth 1 https://github.com/msolecki/founder-osWrote 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/msolecki/founder-os/scope-guard)<a href="https://agentmods.dev/skills/msolecki/founder-os/scope-guard"><img src="https://agentmods.dev/badge/skills/msolecki/founder-os/scope-guard.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 | $0.00033 | $0.01867 |
| Opus 5 | $0.00016 | $0.00933 |
| Sonnet 5 | $0.00007 | $0.00373 |
| Haiku 4.5 | $0.00003 | $0.00187 |
Grade A, and why
scope-guard 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 — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scope Guard
Scope creep does not arrive as a demand. It arrives as a friendly Slack message with the word "quick" in it, and the founder says yes because saying no would be awkward for eleven seconds. This skill makes the eleven seconds cheaper than the hours.
When to use
The moment an ask lands that was not in the proposal — "while you're in there", "one small tweak", "could you also just". Also at every renewal, to count what was absorbed last time and price it.
Inputs
Read first, in order — house rule 1:
pipeline.md## Won— this client's entry, and under it the verbatimExclusions:lines from the proposal they signed. Those lines are your baseline. Read them; do not summarise them.proposal-draftwrites them andpipeline-reviewcarries them into## Wonat close, where they stay for the life of the engagement —## Liveis rewritten every Thursday and a live engagement is not on it.offer.md— the offer boundary: what this company sells at allclients/<client>.md— what has already been absorbed on this engagement, and the running count
If the client has no ## Won entry, or the entry carries no Exclusions:
lines, stop. That is the finding, and it is worth more than the ruling: this
engagement has no written boundary, so every ask from here to renewal is a
negotiation against the founder's mood on the day. Hand to the Pipeline Coach
— proposal-draft owns that gap.
A count is not a list. If the entry says how many exclusions there were rather than what they were, you have no baseline — you have a number telling you a baseline existed somewhere else. Treat it as the empty case and hand it over.
Beliefs
- An exclusion earns its value when it is read aloud, not when it is produced. Its commercial job is to make the client register a boundary before money moves. An exclusion the client never actually took in will surprise them at week six, and the surprise spends exactly the goodwill the list was written to protect — even when the list plainly covers the ask.
- Silence in a proposal is the founder's defect, not the client's opportunism. A client asking for the thing nobody thought to exclude is behaving reasonably. Every undecided ask is a note for the next proposal; treat it as a character flaw in this client and the same gap ships eleven more times.
- "Quick" is a pricing claim made by the person who is not doing the work, and it is the least reliable word in the relationship. It does not mean the estimate is small. It means there is no estimate.
- Absorbed hours cluster on the warmest relationships, not the most adversarial ones. Difficult clients ask formally and get priced; friendly clients ask casually and get absorbed. Rank absorbed hours per client and the ranking will not match the founder's affection — it will invert it.
What ships with it
1 file 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.
- 4d ago First seen · 153 lines · 33 tokens per session scan A dfd1c18a29b2
scope-guard is a skill published in the GitHub repository msolecki/founder-os (4 stars, last pushed yesterday), licensed MIT. It adds 33 tokens to every session and 1,867 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
workthreads
SpecStory Workthreads - a weekly work-thread rollup across a team's repos from SpecStory coding histories (any agent - Claude Code, Codex, Cursor, Gemini, and more). It groups the window's sessions into threads of work per project and labels each new / open / recently closed, so a lead sees what shipped, what is still…
story-readiness
Validate that a story file is implementation-ready. Checks for embedded GDD requirements, ADR references, engine notes, clear acceptance criteria, and no open design questions. Produces READY / NEEDS WORK / BLOCKED verdict with specific gaps. Use when user says 'is this story ready', 'can I start on this story', 'is…
flow-next-tracker-sync
Project a flow-next spec to a tracker issue (Linear, GitHub, GitLab, Jira) and reconcile two-way. Use when asked to sync to a tracker. NOT plan-sync.
loop
Full execution protocol for MODE: LOOP — the compound-engineering loop: brainstorm → plan → build → review → improve, iterating under defense-in-depth stop conditions with generator/critic separation, durable resumable state, and mandatory compounding learning capture. Loaded on demand by the architect when the loop…
subagent-delegation
Canonical protocol for delegating GSD work to native Antigravity subagents — when to delegate, how to invoke, workspace isolation modes, and the inline fallback for older IDE versions.
projects
List all managed projects with status, branch, open PRs, and open issue counts — portfolio-level view.