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 inureyes/gyeol --skill gyeol-capturegit clone --depth 1 https://github.com/inureyes/gyeolWrote 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/inureyes/gyeol/gyeol-capture)<a href="https://agentmods.dev/skills/inureyes/gyeol/gyeol-capture"><img src="https://agentmods.dev/badge/skills/inureyes/gyeol/gyeol-capture/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/inureyes/gyeol/gyeol-capture"><img src="https://agentmods.dev/badge/skills/inureyes/gyeol/gyeol-capture.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 | $0.00066 | $0.00611 |
| Opus 5 | $0.00033 | $0.00305 |
| Sonnet 5 | $0.00013 | $0.00122 |
| Haiku 4.5 | $0.00007 | $0.00061 |
Grade A, and why
gyeol-capture 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.
How it starts
The opening of the file, as written. The whole thing — 34 lines — stays where its author put it; the contents beside it link to each section on GitHub.
gyeol Run Capture
Why this exists: delegated runs execute through hands (subagents, a second harness) whose experience never reaches episodic memory by itself. The 2026-07 retrospective found 145 unrecorded delegated sessions and 8 false "not started" state claims in _recent.md; recording at the moment the run ends, when context is richest, is the fix. This step has the same status as technical reports: part of the run, not optional post-work.
This skill is wired non-invasively: the gyeol session instructions say to follow it at run end when it is installed and to skip when it is not. No orchestration skill references it explicitly, so treat the end of any orchestrated run as the trigger, without waiting to be asked.
When
- After an orchestrated run's final announcement/summary, before ending the turn.
- On abort or interruption: capture what completed so far and mark the cut point explicitly.
- SKIP only when running as a dispatched unit inside a wave/orchestration run (e.g. a wave-runner unit). The orchestrator records the whole run; per-unit entries would duplicate it.
What to write
- Append one compressed section to
$GYEOL_HOME/memory/episodes/daily/{YYYY-MM-DD}.md(create the file withdate/sessionsfrontmatter if missing):- Heading:
## {repo}: {command as invoked} ({outcome}). - 3-8 bullets: units → PRs with merge state, key decisions, defects found, deviations from the plan, open follow-ups.
- Verify merge/close states with
ghat write time; do not write from recall. A state claim ("merged", "not started", "waiting") written from memory can be false before the day ends, so prefer resolvable facts (PR numbers, states) over judgments. - Do not invent introspection for delegated work. Record facts and the reports received; mark reconstructed gaps rather than filling them.
- Heading:
- Update
$GYEOL_HOME/memory/episodes/_recent.md:- Add a one-line Daily Index entry pointing at the daily log.
- Reconcile Still Open: add new unresolved items (tagged with source date), remove items this run resolved.
- Refresh the
last_updatedfrontmatter.
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 · 34 lines · 66 tokens per session scan A 86f0376c3246
gyeol-capture is a skill published in the GitHub repository inureyes/gyeol (51 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 66 tokens to every session and 611 once invoked, about $0.0003 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
media-ingest
Ingest video, audio, PDF, book, screenshot, and GitHub repo content into the brain. Multi-format handling with entity extraction and backlink propagation. Covers video-ingest, youtube-ingest, and book-ingest subtypes.
mem0-oss-to-platform
Plan and then execute a migration of a project from the mem0 open-source / self-hosted SDK (the local Memory class) to the mem0 Platform / hosted / managed SDK (the MemoryClient class). Use this whenever a developer wants to move, switch, or migrate their mem0 usage off OSS/self-hosted to the hosted API — e.g.…
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…
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…
shellm
Reference for the shellm system — recursive LLM shell, identity management, memory, skills, trajectory, and all CLI tools. Use when working on shellm itself, debugging agent behavior, or understanding how the pieces fit together.