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 commands/llama-farm/llamafarm/plangit clone --depth 1 https://github.com/llama-farm/llamafarmWhat 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 | $0.00000 | $0.01141 |
| Opus 5 | $0.00000 | $0.00571 |
| Sonnet 5 | $0.00000 | $0.00228 |
| Haiku 4.5 | $0.00000 | $0.00114 |
Grade A, and why
plan 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 yesterday.
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 — 141 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Implementation Planning Playbook
Turn your completed research into a minimal, verifiable change plan. This document helps you define scope, sequence work, and specify how you’ll prove success before writing code.
Inputs
- The research note in
thoughts/shared/research/<issue_id>_<slug>.md - Clarifications from the user for any uncertainties identified in research
- Awareness of repo guidelines in
AGENTS.mdand project conventions - Answers to the “Known unknowns” listed in the research note (resolve with the user first)
Outputs
- A plan document at
thoughts/shared/plans/<issue_id>_<slug>-plan.mdthat includes:- Clear scope and acceptance criteria
- File-level change list and impact analysis
- Test and verification plan (commands you’ll run)
- Risk/rollout/rollback strategy
- Docs and config update plan
- Estimated effort and sequencing
Planning Workflow
- Confirm problem and success criteria from the research note.
- Identify impacted projects and targets (prefer workspace graph/tools when available).
- Map entry points and contracts to be changed; list interfaces/CLIs/APIs affected.
- Draft the smallest viable change list; postpone refactors.
- Define the test strategy and exact commands you will run to verify.
- Specify risk controls: flags, canaries, and rollback steps.
- List required updates to docs and configuration.
- Sequence tasks and add rough estimates; call out external dependencies.
- Share the plan; adjust based on feedback before coding.
Change List: plan by files and boundaries
For each change, capture:
- File(s)/Module(s): paths to create/edit/remove
- Change summary: what will be added/modified and why
- Contracts: CLI flags, API schemas, function signatures, data shapes
- Dependencies: services, tasks/targets, feature flags, migrations
- Blast radius: who calls this code; risks and mitigations
Example structure:
- server/api/items.py: add GET /items/{id} handler returning X schema
- server/services/items.py: implement fetch_item with validation Y
- cli/cmd/items.go: add `lf items get` to call server endpoint
- docs/website/...: add usage docs and examples
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.
- yesterday First seen · 141 lines · 0 tokens per session scan A a0d7159ad7f2
plan is a command published in the GitHub repository llama-farm/llamafarm (837 stars, last pushed 2mo ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,141 tokens. 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.
Other commands, from other repositories
sync-go-engine
Synchronize the Go side — the engine runtime — with the current contract/workflow.yaml. The TypeScript side has its own sync command (/sync-ts-domain); this one does not touch ts/. Use this after the contract YAML has been edited (or pulled in from fh-backend) to add/remove/change nodes or their argument shapes.
sync-ts-domain
Synchronize the TypeScript side — workflow-core domain layer (and any workflow-builder knock-on edits) — with the current contract/workflow.yaml. The Go engine side has its own sync command; this one does not touch go/. Use this after the contract YAML has been edited (or pulled in from fh-backend) to…
audit
Perform a thorough audit of the React canvas in ts/workflow-builder/src/ against five categories. Use Agent subagents to explore categories in parallel where possible.
new-node
Author a new workflow node end-to-end — interview intent, design the contract schema, then reconcile both the TS and Go sides.
sync-translations
Audit the i18n trees for translation completeness across all locales. The hard, error-prone part — which node/param/edge/channel/memory/model descriptions still lack translations — is derived deterministically by a script; the rest are agent checks.
model-check
Reassess the current session and recommend the right model and effort level.