config

A command for viewing or changing OpenArche configuration, including task orchestration and knowledge-search settings.

In plain words
What is it for?
Use it to check configuration, change orchestration or embedding settings, and identify invalid configuration before OpenArche runs.
Why use it?
It gives one place to inspect the settings that control how OpenArche handles tasks and stores searchable knowledge.

Command

Install

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.

agentmods
npx agentmods add commands/done-0/openarche/config
Clone the repo
git clone --depth 1 https://github.com/Done-0/openarche
Per session 7 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 724 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00007 $0.00724
Opus 5 $0.00003 $0.00362
Sonnet 5 $0.00001 $0.00145
Haiku 4.5 $0.00001 $0.00072

Measured 2d ago against content hash ca4a05e45486, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

config 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.

commands/config.md · 83 lines

What it actually says

Use this command to inspect or change the single source of truth for OpenArche configuration.

Get the config file path:

node -e "const os=require('os'),path=require('path');console.log(path.join(os.homedir(),'.claude','openarche','config.json'));"

Read that file.

  • If it does not exist, tell the user to run /openarche:setup first.
  • If it exists but is invalid JSON or does not match the required config shape, explain that the config is invalid and must be corrected before OpenArche can run reliably. Do not pretend the current file is usable.

When showing the current config, group it by section and include the config path:

Orchestration

  • orchestration.autoInject
  • orchestration.persistAfterFirstToolUse — one of false, write_only, or execute_or_write Production default is write_only, so read-only analysis does not materialize a persisted task session.
  • orchestration.readOnlyCommands
  • orchestration.explicitSessionCommands
  • orchestration.injectOnlyIntentThreshold
  • orchestration.materializeIntentThreshold

Knowledge

  • knowledge.embedding.provider
  • If provider is local, show knowledge.embedding.localModel
  • If provider is remote, show knowledge.embedding.remoteModel and knowledge.embedding.remoteBaseUrl
  • Never print the full knowledge.embedding.remoteApiKey; show it as redacted or masked
  • knowledge.retrieval.threshold
  • knowledge.retrieval.topK
  • knowledge.retrieval.maxInjectChars
  • knowledge.extraction.model
  • knowledge.extraction.minQualityScore
  • knowledge.extraction.captureConcurrency

Execution

  • execution.isolationStrategygit-worktree or git-branch
  • execution.baseRef

Validation

  • validation.browser.enabled
  • validation.browser.captureDomSnapshot
  • validation.browser.captureScreenshot
  • validation.browser.captureNavigation

Observability

  • observability.enabled
  • observability.logs
  • observability.metrics
  • observability.traces

Review

  • review.localSelfReview
  • review.localAgentReview
  • review.cloudAgentReview
  • review.repairLoops

Maintenance

  • maintenance.qualitySweep
  • maintenance.driftSweep

When editing:

  • Keep orchestration.materializeIntentThreshold greater than or equal to orchestration.injectOnlyIntentThreshold.
  • Only write fields that belong to the selected embedding provider.
  • If switching to local, remove remote-only embedding fields.
  • If switching to remote, remove localModel and require remoteModel, remoteApiKey, and remoteBaseUrl.
  • Keep the final file fully valid. Do not leave partial config.
  • If the user selects local, remind them that the first successful embedding call may need network access to download model files.
  • If the user edits remoteApiKey, accept the new value but do not echo the full key back in the response.

After writing the file back:

  • Tell the user exactly which fields changed.
  • If any embedding provider or embedding model field changed, remind the user:

You changed the knowledge embedding settings. Run /openarche:knowledge-reindex to rebuild the vector index.

Changes

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.

  1. 2d ago First seen · 83 lines · 7 tokens per session scan A ca4a05e45486

Subscribe to this mod's changes

config is a command published in the GitHub repository Done-0/openarche (12 stars, last pushed 4mo ago), licensed MIT. It adds 7 tokens to every session and 724 once invoked, about $0.0000 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 2026-08-30.