Getting it into your agent
There is no command for this one: it runs only inside a plugin, and the catalogue could not identify which plugin ships it. The source is linked below.
Wrote 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/lossless-group/lossless-agent-skills/search-lossless-corpus)<a href="https://agentmods.dev/skills/lossless-group/lossless-agent-skills/search-lossless-corpus"><img src="https://agentmods.dev/badge/skills/lossless-group/lossless-agent-skills/search-lossless-corpus/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/lossless-group/lossless-agent-skills/search-lossless-corpus"><img src="https://agentmods.dev/badge/skills/lossless-group/lossless-agent-skills/search-lossless-corpus.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 | $0.00230 | $0.02971 |
| Opus 5 | $0.00115 | $0.01486 |
| Sonnet 5 | $0.00046 | $0.00594 |
| Haiku 4.5 | $0.00023 | $0.00297 |
Grade A, and why
search-lossless-corpus 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 12d 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 — 160 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Search the Lossless Corpus
The Lossless Group keeps a living corpus of its own decisions, shipped work, prior sessions, and tool-failure history in a local Chroma database wired into Claude Code via the chroma MCP server. Most questions a user asks about prior work can be answered from this corpus rather than training data. Don't paraphrase from training data when the corpus could answer.
This skill is the discipline that makes the corpus load-bearing. Without it, Chroma sits there full of high-signal records that nobody queries. With it, every "what did we decide about X" question becomes grounded recall — with citations the user can verify in seconds.
When to use this skill
Trigger when the user's question takes any of these shapes:
- Decisions — "what did we decide about X", "why did we choose X over Y", "what's our stance on X"
- Shipping — "when did we ship X", "what shipped in May", "did we already do X"
- History — "what did we talk about last week", "in a prior session we…", "resume where we left off on X"
- Failures — "has this errored before", "how did we fix X last time", "when did Bash with git rebase fail"
- Location — "where did we put the X spec", "which repo's changelog mentions Y"
Trigger also when the user pastes a question that touches Lossless conventions (changelog format, context-v structure, branch-tier model) — the skill canon may have moved since your training cutoff; the corpus reflects current truth.
The four collections
The MCP server exposes them via tools named mcp__chroma__chroma_query_documents, mcp__chroma__chroma_list_collections, mcp__chroma__chroma_get_documents, etc. Each collection is keyed to a different question shape:
| Collection | Granularity | Best for | Key metadata fields |
|---|---|---|---|
context-vigilance-corpus |
Section-level chunks of context-v/ files |
Architecture, specs, plans, conventions, "what do we believe about X" | source_repo_slug, source_path, kind, chunk_heading, fm_title, fm_status, fm_tags |
lossless-changelog |
Whole changelog entries | "What shipped about X and when", cross-repo | source_repo_slug, source_path, fm_date, fm_title, fm_lede, fm_tags |
claude-code-sessions |
One document per message turn | "What did we decide about X", resume context | session_id, project_path, turn_role, timestamp, git_branch |
claude-code-tool-traces |
One document per tool invocation | "When did X tool fail and how did we recover" | tool_name, is_error, session_id, timestamp, git_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.
- 12d ago First seen · 160 lines · 230 tokens per session scan A bbb787788bc1
search-lossless-corpus is a skill published in the GitHub repository lossless-group/lossless-agent-skills (4 stars, last pushed 19d ago), licensed MIT. It adds 230 tokens to every session and 2,971 once invoked, about $0.0011 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
memory-audit
Audit MEMORY.md against the memory discipline — oversized sections, settled multi-session patterns that belong in knowledge/concepts/, stacked chronicle blocks, stale entries. Produces a move plan as a table for approval, then executes the approved moves atomically. Use when the SessionStart hook reports a tripped…
close-session
An end-of-session workflow that reviews new lessons, asks before promoting repeated patterns, refreshes memory files, and writes a handoff.
aictx
Use this skill when working in a repository that uses AICTX for repo-local operational continuity. Prefer AICTX MCP tools when available, fall back to CLI commands, resume before substantial work, and finalize factual continuity after work.
tour
An interactive walkthrough of the Memory Kit system using the files in the current project. It explains how session handoffs, task backlogs, and stored project memory fit together.
journal-optimizer
Guided database pruning and optimization workflows for memory-journal-mcp. Uses importance scores, relationship density, and entry metadata to identify low-value entries for safe soft-deletion. Includes dry-run previews, backup gates, and revert guidance. Use when the user says "clean up the database", "optimize…
init
Initialize structured context for this project. Scaffolds .scs/ with all 11 concern bundles, scans the project, and recommends where to start.