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/greppable/greppable-cc-plugin/traversing-gdlm-memorynpx skills add greppable/greppable-cc-plugin --skill traversing-gdlm-memorygit clone --depth 1 https://github.com/greppable/greppable-cc-pluginWrote 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/greppable/greppable-cc-plugin/traversing-gdlm-memory)<a href="https://agentmods.dev/skills/greppable/greppable-cc-plugin/traversing-gdlm-memory"><img src="https://agentmods.dev/badge/skills/greppable/greppable-cc-plugin/traversing-gdlm-memory.svg" alt="Measured on agentmods" 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 | $0.00145 | $0.01907 |
| Opus 5 | $0.00072 | $0.00954 |
| Sonnet 5 | $0.00029 | $0.00381 |
| Haiku 4.5 | $0.00015 | $0.00191 |
Grade A, and why
traversing-gdlm-memory scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 194 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Memory Quick Reference
Available Memory Files
!bash -c 'find docs/gdl/memory/active -name "*.gdlm" -maxdepth 1 2>/dev/null | while read f; do count=$(grep -c "^@memory" "$f" 2>/dev/null || echo 0); echo "- $(basename "$f"): $count records"; done'
Format
@memory|id:{id}|agent:{agent}|subject:{subject}|tags:{t1,t2}|detail:{text}|confidence:{level}|anchor:{concept}|ts:{timestamp}
Each memory is one line. Each field is self-describing (key:value). Latest timestamp wins for corrections.
Required Fields
| Field | Description |
|---|---|
id |
M-{agent}-{seq} unique identifier |
agent |
Agent that authored the memory |
subject |
Primary topic (flat) |
detail |
The knowledge content |
ts |
ISO timestamp |
Optional Fields
| Field | Description |
|---|---|
tags |
Comma-separated retrieval tags |
confidence |
high, medium, low |
relates |
Links to other memory IDs (typed: supersedes~M-001) |
type |
observation, decision, preference, error, fact, task, procedural |
anchor |
Concept anchor reference |
status |
corrected, deleted, compacted |
Three Tiers
| Tier | Path | Purpose |
|---|---|---|
| Active | memory/active/*.gdlm |
Current working memory (read/write) |
| Archive | memory/archive/*.gdlm |
Compacted summaries (read-only) |
| History | memory/history/{period}/*.gdlm |
Full originals (forensic only) |
Concept Anchors
Two-grep similarity search via memory/anchors.gdlm:
# Step 1: Find anchor by keyword
grep "authentication\|oauth" memory/anchors.gdlm
# → @anchor|concept:auth-security|scope:authentication,authorization,oauth,...
# Step 2: Find memories by anchor
grep "anchor:auth-security" memory/active/*.gdlm
Grep Patterns
# By subject
grep "subject:GL_ACCOUNT" memory/active/*.gdlm
# By agent
grep "agent:sf-042" memory/active/*.gdlm
# By tag
grep "tags:.*snowflake" memory/active/*.gdlm
# By anchor concept
grep "anchor:data-quality" memory/active/*.gdlm
# By type
grep "type:decision" memory/active/*.gdlm
# Latest version of a memory
grep "id:M-sf-042-001" memory/active/systems.gdlm | tail -1
# High-confidence only
grep "confidence:high" memory/active/*.gdlm
# Related memories
grep "relates:.*M-sf-042-018" memory/active/*.gdlm
# Search archive when active doesn't answer
grep "subject:GL_ACCOUNT" memory/archive/*.gdlm
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.
- 5d ago First seen · 194 lines · 145 tokens per session scan A 3b8a32c26115
traversing-gdlm-memory is a skill published in the GitHub repository greppable/greppable-cc-plugin (3 stars, last pushed 4mo ago), licensed MIT. It adds 145 tokens to every session and 1,907 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
consolidating-notes
Merge an accumulated pile of agent notes, memories, or scratch documents into repository knowledge — freezing the input, synthesizing into a separate output, diffing, and routing each surviving entry to a doc, a guard, a subtree CLAUDE.md, or the small remainder that really is user-scoped. Also covers pruning docs and…
rk-pull
Pull knowledge base from remote and merge into local. Fetches remote state, Agent compares diffs and merges (alias LRU merge for index.md, union dedup for registry.md, newer-cached-date wins for docs). Does NOT push — local only. Use when user says 'pull knowledge', 'sync from remote', 'rk-pull', or on a new machine…
rk-push
Push local knowledge base to the remote repository. Always fetches remote first, Agent compares diffs and merges (alias LRU merge for index.md, union dedup for registry.md, newer-cached-date wins for docs), then commits and pushes. Retries once on push rejection. Use when user says 'push knowledge', 'sync to remote'…
rk-update
Incrementally update a codebase knowledge base by checking git commit history. Finds changed files since last update, regenerates only affected documentation, and updates the index. Auto-pulls from remote before updating and auto-pushes after if remote is configured. Use when user says 'update knowledge base'…
rk-memo
Save a piece of personal knowledge to the user's personal knowledge base (personal project). No git repo needed — stores anything the user knows by heart: commands, tips, best practices, personal workflows. Searchable via rk-search, synced via rk-push/pull. Use when user says 'remember this', 'save this note'…
memory-recall
Search and recall relevant memories from past sessions via memsearch. Use when the user's question could benefit from historical context, past decisions, debugging notes, previous conversations, or project knowledge -- especially questions like 'what did I decide about X', 'why did we do Y', or 'have I seen this…