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 ReclaimLLM/RCLM --skill reclaimllm-replaygit clone --depth 1 https://github.com/ReclaimLLM/RCLMWrote 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/reclaimllm/rclm/reclaimllm-replay)<a href="https://agentmods.dev/skills/reclaimllm/rclm/reclaimllm-replay"><img src="https://agentmods.dev/badge/skills/reclaimllm/rclm/reclaimllm-replay/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/reclaimllm/rclm/reclaimllm-replay"><img src="https://agentmods.dev/badge/skills/reclaimllm/rclm/reclaimllm-replay.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 | $0.00184 | $0.01935 |
| Opus 5 | $0.00092 | $0.00967 |
| Sonnet 5 | $0.00037 | $0.00387 |
| Haiku 4.5 | $0.00018 | $0.00194 |
Grade A, and why
reclaimllm-replay 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 11d 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 — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ReclaimLLM Replay
Replay reproduces RCLM's compression claim against the user's own captured sessions, on their machine, in the same units the claim was published in: text tool-result tokens removed. It is strictly read-only — it never writes to the database, never re-executes a historical command, and never calls a model. The claim being verified is narrow and must be stated exactly this way when relaying results:
a replayed tool-result token reduction — not a claim about total model input, billing, or live user savings.
Core Pattern
- Always call
replay_eligibilityfirst. It is a cheap, metadata-only check — no blob fetch — and answers "is this worth replaying" before any real computation runs. - If eligible, call
replay_session(one session) orreplay_corpus(a filtered window) depending on what the user asked about. - Only call
replay_comparewhen the user explicitly wants multiple mechanism configurations compared against the same corpus in one call. - Narrate the result plainly. Do not decide for the user whether the number
is "good" — state it, state the funnel, state
cannot_tell_you.
Tool Routing
replay_eligibility(session_id?, days?, source?, model_family?, project?, session_category?, limit?, min_turns?, min_tool_calls?)— call this before either tool below. Passsession_idto check one session; omit it to check a corpus window. Returnseligible, the failing constraint if not, and the funnel (considered/eligible/excluded).replay_session(session_id?, mechanisms?, min_turns?, min_tool_calls?)— one session. Defaults to the caller's most recent complete session.mechanismsdefaults to all three (range_cache,shell_compaction,hash_dedupe); pass a subset only if the user wants to isolate one mechanism.replay_corpus(days=30, source="all", model_family?, project?, session_category?, mechanisms?, limit?, min_turns?, min_tool_calls?)— a filtered set of the caller's own sessions.daysis an exact rolling ingestion window measured oningested_at. Forsource="codex", stored models must start withgpt-orcodex-.limitis the target fully eligible session count. Replay fetches up to four times that many recentsessionrecords (capped at 100), applies both eligibility tiers in order, and stops atlimiteligible sessions or scan exhaustion.replay_compare(days=30, source="all", ..., configs?, min_turns?, min_tool_calls?)— same corpus, multiple mechanism sets, one call, so bundles stay attributable. Only use when the user is comparing configurations, not for a single verification.
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.
- 11d ago First seen · 151 lines · 184 tokens per session scan A 34216dd67283
reclaimllm-replay is a skill published in the GitHub repository ReclaimLLM/RCLM (0 stars, last pushed 12d ago), licensed Apache-2.0. It adds 184 tokens to every session and 1,935 once invoked, about $0.0009 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…