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/zeikar/hyperclaude/hyper-recapnpx skills add zeikar/hyperclaude --skill hyper-recapgit clone --depth 1 https://github.com/zeikar/hyperclaudeWhat 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 | $0.00135 | $0.04795 |
| Opus 5 | $0.00068 | $0.02397 |
| Sonnet 5 | $0.00027 | $0.00959 |
| Haiku 4.5 | $0.00014 | $0.00479 |
Grade A, and why
hyper-recap 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 3d 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 — 165 lines — stays where its author put it; the contents beside it link to each section on GitHub.
hyper-recap
Cycle-completion recap. An AI-run cycle leaves the codebase changed but the human out of the loop: the request, the judgments taken, the alternatives rejected, and the critic rounds are all in the AI's head or scattered across gitignored artifacts. That is cognitive debt — without a write-up a human can't reconstruct why and how the AI built what it built. This skill collapses one completed cycle into a single human-readable recap under .hyperclaude/recaps/.
Claude-only — no Codex, no agent dispatch. It joins the non-Codex-spawning set alongside hyper-interview, hyper-memory, and hyper-setup. It never calls the bridge and never dispatches an agent — it only reads .hyperclaude/ artifacts + in-session context and writes one markdown file.
The recap unit is ONE completed detailed plan / milestone (a plan archived under .hyperclaude/plans/done/). Epic-level rollup across milestones is a v1 non-goal. This is NOT a PR draft — it explains the cycle to a human, it does not propose or format changes for merge.
When to use
- User typed
/hyperclaude:hyper-recap [path|slug]. - A cycle just completed — the plan ran to completion (
implement/code-reviewdone) and was archived under.hyperclaude/plans/done/— and the why/how should be captured. - hyper-auto invokes this skill automatically as its terminal step on a clean composed exit (the sole auto-run path; standalone it stays on-demand).
When to skip
- Mid-cycle. v1 recaps ONLY completed cycles — there is no mid-cycle partial-recap mode. If the target plan is still active (not yet archived), stop and say so (see Target resolution's completion gate).
- No completed plan exists. Zero
.hyperclaude/plans/done/*.md→ report "nothing to recap", write NOTHING, stop.
Invocation argument
Invocation argument: $ARGUMENTS
Target resolution
Every branch resolves to a COMPLETED plan under .hyperclaude/plans/done/. Classify $ARGUMENTS:
- contains
/OR ends in.md→ path branch - a bare non-empty token → slug branch
- empty → no-arg branch
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.
- 3d ago First seen · 165 lines · 135 tokens per session scan A 5fdb2457298a
hyper-recap is a skill published in the GitHub repository zeikar/hyperclaude (3 stars, last pushed 15d ago), licensed MIT. It adds 135 tokens to every session and 4,795 once invoked, about $0.0007 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
workflow
Professional AI programming assistant with structured workflow (Research -> Ideate -> Plan -> Execute -> Optimize -> Review) for developers.
git-rollback
Interactively rollback Git branch to historical version; lists branches, versions, then executes reset/revert after confirmation.
golden-rss
Use when testing the rss golden build.
design-system-builder
Extract design systems from reference UI images and generate implementation-ready UI design prompts. Use when users provide UI screenshots/mockups and want to create consistent designs, generate design systems, or build MVP UIs matching reference aesthetics.
r3f-best-practices
React Three Fiber (R3F) and Poimandres ecosystem best practices. Use when writing, reviewing, or optimizing R3F code. Triggers on tasks involving @react-three/fiber, @react-three/drei, zustand, @react-three/postprocessing, @react-three/rapier, or leva.
log-error-digest
Analyze log files to troubleshoot errors, identify peak error periods, and produce error clustering, frequency statistics, and time distribution reports. Supports JSON, syslog, and Nginx formats with automatic detection. Use when a user uploads a .log file and asks to analyze errors, find patterns, debug issues, or…