Borrowing it
Nothing to install: this file belongs to CorridorTech/PoseCap. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/CorridorTech/PoseCap/main/.agents/skills/ad-handoff/SKILL.mdgit clone --depth 1 https://github.com/CorridorTech/PoseCapWrote 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/corridortech/posecap/ad-handoff)<a href="https://agentmods.dev/skills/corridortech/posecap/ad-handoff"><img src="https://agentmods.dev/badge/skills/corridortech/posecap/ad-handoff/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/corridortech/posecap/ad-handoff"><img src="https://agentmods.dev/badge/skills/corridortech/posecap/ad-handoff.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 178 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00135 | $0.03828 |
| Opus 5.5 | $0.00054 | $0.01531 |
| Sonnet 5.5 | $0.00027 | $0.00766 |
| Haiku 4.5 | $0.00014 | $0.00383 |
Grade A, and why
ad-handoff 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 2mo 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 — 198 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<background_information> Process scaffold for compacting a long or near-exhausted session into a handoff document. Output is a single markdown file in the OS temp directory; the next session reads it and continues. Helper, not blocker — the senior engineer keeps decision authority over what survives the compaction.
The skill exists because long sessions accumulate inherited bias (WORKFLOW §12 — "almost right" failures compound) and because session-window pressure forces ad-hoc summarisation that loses the parts the next agent actually needs. Handoffs replace ad-hoc summarisation with a structured artifact.
Codex subagents and sessions do not share the parent conversation history by magic, and /clear discards local context. Handoff via the OS temp dir is the durable bridge between Codex sessions or explicit subagent work — treat it as load-bearing, not optional. Codex auto-trigger on description keywords is less mature than Claude Code's; if auto-invocation does not fire when the user mentions handing off, compacting, or /clear, invoke this skill manually.
</background_information>
Route elsewhere when:
- The work is finishable in the current turn → finish it; do not pre-emptively hand off.
- The handoff target is a specific artifact (PRD / spec / task / ADR) → write that artifact via the matching
ad-*skill instead. Handoffs do not replace persistent decision records. - The user wants to commit work-in-progress →
ad-commitor finish the change first.
Step 1 — collect live state. Do not duplicate content that already lives in repo artifacts; reference those by path.
Capture:
- Working tree —
git status --short, current branch (git branch --show-current), divergence (git log --oneline @{upstream}..HEADif a remote is set). - In-progress diff —
git diff+git diff --staged. If the combined diff is >300 lines, summarise per-file and tell the next agent to re-read it withgit diff. - Open artifacts — files touched but uncommitted; task file under
doc/tasks/driving the work (if any); spec underdoc/specs/it implements (if any); ADR being drafted underdoc/adr/(if any). - Unresolved decisions — questions the user posed that you have not answered; questions you posed that the user has not answered. One bullet per open question, with a recommended-answer line where you have one.
- Recent errors / hook failures — verbatim, last error message + which command produced it.
- Repo hygiene —
git worktree list,git branch --merged <default>,git branch --no-merged <default>, stray untracked files at the repo root. Delete the merged branches:git branch -d <name>refuses anything not fully merged, so git itself is the guard, and a merged branch's commits survive on the default branch and in the reflog — the blast radius WORKFLOW §7 asks you to match is near zero. Run it and report what was deleted. Everything else is reported with its command and never run: unmerged branches (including squash-merged ones, which git cannot distinguish from genuinely unmerged), worktrees, and stray files getgit branch -D <name>/git worktree remove <path>written next to them. Those are the destructive cases —-Ddiscards commits. Known limit, tested rather than assumed: squash-merged branches cannot be detected —git branch --merged,git cherryand a three-dot diff all miss them, because squashing rewrites the patch-id. Report as ambiguous; a heuristic that guesses here would authorise deleting real work. - What the next agent should do first — one sentence, imperative.
References, not copies. Spec: doc/specs/0007-foo.md beats pasting the spec. The next agent will read those files fresh.
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.
- 2mo ago First seen · 198 lines · 135 tokens per session scan A b3195ee42aae
ad-handoff is a skill published in the GitHub repository CorridorTech/PoseCap (265 stars, last pushed 3d ago), licensed Apache-2.0. It adds 135 tokens to every session and 3,828 once invoked, about $0.0005 per session on Opus 5.5. 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
knowledge-ingest
Ingest documents, articles, conversations, or supplied text into Knowledge with retained originals, sourced claims, entity updates, source notes, and cross-links; also learn durable lessons from a set of conversations, transcripts, or past sessions (/learn), including the Learn from conversations and Learn from Memory…
mem0-oss-to-platform
Plan, then execute, a migration of a project from the mem0 open-source / self-hosted SDK (local Memory class) to the mem0 Platform / hosted SDK (MemoryClient). TRIGGER when: the developer wants to move, switch, or migrate mem0 off OSS/self-hosted to the hosted API, e.g. "migrate my mem0 setup to the platform", "switch…
forget
Remove a personal fact, preference, relationship detail, topic, or prior event from Muse's active memory and stop existing copies or automations from bringing it back. Use for explicit requests such as 'forget that', 'don't remember this about me', or 'remove that from your memory'. Do not use when 'forget it' merely…
Cortex
LifeOS memory, prior-session recall, and learning capture. USE WHEN capture this learning, document this learning, what did we learn recently, recent learnings, cortex, knowledge, knowledge base, search knowledge, what do we know about, archive, harvest, knowledge status, develop note, add to knowledge, add a people…
memory
Use when the user asks to remember, recall, forget, update, search, or inspect durable OpenSquilla memory, including profile facts in USER.md and long-term notes in MEMORY.md or memory//.md.
ha-data-stores
Map of Hope Agent's local data stores and safe read-only query workflow. Use when the user asks where Hope Agent stores data, wants to inspect sessions/messages/memory/logs/background jobs/knowledge indexes/settings, asks the model to query local app data, or debugging requires checking persisted state. Trigger…