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 reeswrites/exo --skill pick-up-a-projectgit clone --depth 1 https://github.com/reeswrites/exoWrote 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/reeswrites/exo/pick-up-a-project)<a href="https://agentmods.dev/skills/reeswrites/exo/pick-up-a-project"><img src="https://agentmods.dev/badge/skills/reeswrites/exo/pick-up-a-project/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/reeswrites/exo/pick-up-a-project"><img src="https://agentmods.dev/badge/skills/reeswrites/exo/pick-up-a-project.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00104 | $0.01214 |
| Opus 5 | $0.00052 | $0.00607 |
| Sonnet 5 | $0.00021 | $0.00243 |
| Haiku 4.5 | $0.00010 | $0.00121 |
Grade A, and why
pick-up-a-project 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 9d 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pick up a project
The workshop tools hold prose and metadata about the owner's repos: what each one claims to be, what was committed and when, what documents it carries, and what is visibly unfinished in it. No source code is in this store — subjects, paths and marker text only. Never imply you read the code.
The failure this skill exists to prevent is reading a stale pointer as current
state. project_open is kind: pointer, and pointers go stale silently.
Needs
| tool | without it |
|---|---|
projects |
you cannot see the shape of the workshop; ask which repo they mean |
project_activity |
the dated work log is gone — do not guess when something died |
project_open |
you can still say what a project is; not what is left in it |
project_docs |
you lose the reasoning; report state without the why |
drafts |
skip the half-written prose behind a repo |
open_threads |
skip the unclosed questions; it takes no filter, so it is all or nothing |
notes_on, whats_relevant |
skip the notes join |
Read status as heat, not as judgement
projects grades each repo by how recently it was committed to — active, warm,
stalled, dormant. That is a measurement of heat and nothing else.
- Dormant is not failed. Everything that simply shipped is dormant. So is everything finished, everything archived, and everything that never needed another commit.
- Deliberately shelved is not the same as gone cold. Where a repo is filed under a group meaning hiatus or similar, that filing is a decision the owner made, not an inference. Report it as their decision.
- Call
projectswith no arguments first. It ranks by commits in the recent window rather than by last-commit date, which is the honest default: half a set of repos can be initialised in one week, and sorting by recency floats a one-commit import above a year of real work.
Reconstruct where they left it, in this order
project_activityfor the repo. Commit subjects, dated. This is the closest thing here to a work log, and unlike a note it cannot be stale — a commit is written at the moment of the work. Read the last ten before you say anything about what the project is in the middle of.project_docs. READMEs, glossaries, architecture decision records, plan documents. This is where a project states what it is for. An ADR is the owner arguing with themselves and recording who won — the same category as their notes and usually more decided, so read it before proposing anything it already settled. Search by topic across repos to find which project answered a question; pass a repo to read what it says about itself.project_open. TODO and FIXME markers, unchecked plan items, uncommitted files. Treat this as a trail, not a backlog. Nobody prunes markers, so an old one may name work that was done another way, or a decision that was reversed. Cross-check every marker you plan to surface againstproject_activityaround and after its neighbourhood, and say when you cannot tell whether it still stands. An uncommitted file is the exception — that is live state, and it is usually the single most useful thing to report.drafts(topic:)andopen_threads. The half-written prose and the unclosed questions the repo itself does not carry.draftstakes a topic;open_threadstakes no arguments at all, so it returns the whole list and you filter it by reading. Do not claim you queried it for this project.
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.
- 9d ago First seen · 93 lines · 104 tokens per session scan A bb4c7519ba0c
pick-up-a-project is a skill published in the GitHub repository reeswrites/exo (0 stars, last pushed 4d ago), licensed MIT. It adds 104 tokens to every session and 1,214 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 2026-09-04.
Other skills, from other repositories
recipe-create-meet-space
Create a Google Meet meeting space and share the join link.
atmos-config
Atmos root configuration: atmos.yaml discovery, precedence, deep merging, basepath, imports, minimal bootstrap, and routing to narrower Atmos skills.
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…
remove
Remove a deployed framework or addon from the current workspace.
handle-linkedin-connection-request-signal
Use this skill when someone sends a team member an inbound LinkedIn connection request. An inbound request is a deliberate, ACTIVE first-party intent signal — meaningfully stronger than a passive profile view — and it deserves different scoring rules. The skill resolves and enriches the requester, gates them through…