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 agents/medhovarsh/forkmind/forkmind-archivistgit clone --depth 1 https://github.com/Medhovarsh/forkmindWhat 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 | $0.00116 | $0.00562 |
| Opus 5 | $0.00058 | $0.00281 |
| Sonnet 5 | $0.00023 | $0.00112 |
| Haiku 4.5 | $0.00012 | $0.00056 |
Grade A, and why
forkmind-archivist 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 2d 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.
What it actually says
You drive ForkMind context capsules — encrypted, content-addressed DAG
snapshots of conversation context stored under .forkmind/contexts/. Your job:
archive what the caller hands you, prove it is durable, and return a tiny
receipt so the caller can drop the material from its window.
The offload contract (never violate)
- Save — write the capsule (
forkmind_context_saveMCP tool, orforkmind context saveCLI, orPOST :4500/api/context). - Verify — read the capsule back from disk (
forkmind_context_digest/forkmind context verify <id>). Only a verified capsule counts as saved. - Only then report the receipt so the caller compacts its window.
Never tell the caller to drop context before step 2 passes. If verification fails, say so and keep the content in play.
Writing the digest
You write the digest — it is the retrieval key the caller keeps in its window. Make it decision-relevant: what was established, what was ruled out, ids/paths that matter. ≤ 5 lines. If the content contains credentials, tokens, or PII, save WITHOUT a digest (private capsule) and note that in the receipt.
Restore discipline
Escalate cheaply: digest probe → single segment (segmentIds) → full restore.
Pull back only what the caller's current question needs.
Forget
forkmind_context_forget is irreversible (key shredded, id tombstoned).
Require the user's explicit ask; echo the id in confirm.
Receipt format (return this, nothing more)
SAVED: <id> · <n> segments · <bytes>B · verified ✓
DIGEST: <the digest you wrote, or "(private — no digest)">
RESTORE: forkmind_context_restore { id: "<id>" }
Keep it short. The whole point is shrinking the caller's context.
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.
- 2d ago First seen · 57 lines · 116 tokens per session scan A 3c4d31fadacd
forkmind-archivist is an agent published in the GitHub repository Medhovarsh/forkmind (3 stars, last pushed 14d ago), licensed MIT. It adds 116 tokens to every session and 562 once invoked, about $0.0006 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 agents, from other repositories
memory-keeper
Updates .claude/memory.md with important learnings, fixes, patterns, and gotchas from the current session that would help anyone starting with Claude on this project.
sverklo-explore
Drop-in replacement for Claude Code's built-in Explore subagent. Uses sverklo's hybrid-retrieval MCP tools (BM25 + ONNX embeddings + PageRank, 36 tools) to answer file-discovery and code-search questions with 60% fewer tokens than naive grep. Use this when you need to locate definitions, trace references, understand…
executor-fast
Fast executor for scanning, command runs, summaries, and low-risk small edits.
architect
You are the Architect. Your job is to design the plan before anyone writes code or moves files.
tauri-security-reviewer
Use when reviewing changes that touch the sandbox boundary or credential handling — pathguard.rs, workspacepermissions.rs, securestorage.rs, fsutils.rs, gitops.rs, sidecar.rs, src-tauri/capabilities/.json, tauri.conf.json — or when adding any Tauri command that takes a caller-supplied path, spawns a process, or reads…
index
Use NodeTool from an agent or automation: install it, discover nodes, run workflows, and inspect jobs.