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 skills add cofin/flow --skill flow-stategit clone --depth 1 https://github.com/cofin/flowWrote 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/cofin/flow/flow-state)<a href="https://agentmods.dev/skills/cofin/flow/flow-state"><img src="https://agentmods.dev/badge/skills/cofin/flow/flow-state/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/cofin/flow/flow-state"><img src="https://agentmods.dev/badge/skills/cofin/flow/flow-state.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Prompt Injection · line 10 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
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 | $0.00030 | $0.02392 |
| Opus 5.5 | $0.00012 | $0.00957 |
| Sonnet 5.5 | $0.00006 | $0.00478 |
| Haiku 4.5 | $0.00003 | $0.00239 |
Grade A, and why
flow-state 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 1mo 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 — 181 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Flow State
Use this skill as the deterministic state boundary for Flow. The active lifecycle owner chooses a legal operation, supplies a complete typed request, and applies that request literally or refuses it with ordinary file tools. Read the packaged canonical state contract before handling any request. It owns the exact payload schemas, read predicates, fragments, event grammar, and lifecycle effects.
Compound requests require ordered {request, read_set, fragments} child evidence.
Only the task operations listed in the canonical contract compose; validate each
step against its predecessor and stamp the outer transaction identity once.
operations: [create, activate, claim, release, note, discover, block, unblock, checkpoint, close, skip, reopen, revise, reconcile, complete, archive, recover, compound, status]
mutation_request:
required: [flow_id, operation, actor, occurred_at, expected_plan_revision, expected_plan_commit, expected_state_revision, targets, payload]
unknown_fields: refuse
missing_fields: refuse
identity_mismatch: refuse_without_writes
payload_schema: canonical_state_contract
status_request:
required: [operation, flow_id, task_ids]
unknown_fields: refuse
writes: none
target_modes:
create.flow: empty
create.task: one
activate: empty
claim: one
release: one
note: one
discover: one
block: one
unblock: one
checkpoint.task: one
checkpoint.phase: empty
checkpoint.plan: all_tasks_sorted
close: one
skip: one
reopen: one
revise: all_tasks_sorted
reconcile: empty
complete: empty
archive: empty
recover: empty
compound: affected_tasks_sorted
lifecycle_guards:
create.flow: [absent]
create.task: [planned, active]
activate: [planned]
active_allowed: [claim, release, note.normal, discover, block, unblock, checkpoint.task, checkpoint.phase, close, skip, reopen, reconcile, complete, compound]
checkpoint.plan: [planned, active]
revise: [planned, active]
archive: [completed]
status: [planned, active, completed]
recover: [planned, active, completed, removed]
completed_allowed: [status, recover, archive, note.git_note_attachment]
removed_allowed: [recover]
identity_routes:
task_target: [create.task, claim, release, note, discover, block, unblock, checkpoint.task, close, skip, reopen]
all_tasks_then_spec: [checkpoint.plan, revise]
task_targets_then_spec: [compound]
spec_only_empty_targets: [activate, checkpoint.phase, reconcile, complete, archive]
untouched_tasks: may_lag_state_revision
snapshot_effects:
current_target: apply_operation_effects
non_current_target: bounded_summary_only
spec_only: typed_affected_ids_not_operation_targets
roots:
configured: setup_state_or_default
bundle: config_or_default
flow: bundle_specs_flow_id
paths: namespaced_relative_no_symlink_or_escape
result_union:
additional_fields: forbidden
keyset: [outcome, operation, flow_id, operation_id, plan_revision, plan_commit, state_revision, targets, journal, evidence, refusal]
outcome_enum: [committed, replayed, rolled_back, recovery_required, contended, refused, status]
field_types:
operation: contract_operation_or_null
flow_id: non_empty_flow_id_or_null
operation_id: canonical_operation_id_or_null
plan_revision: integer_at_least_one_or_null
plan_commit: lowercase_7_to_40_hex_or_null
state_revision: non_negative_integer_or_null
targets: unique_sorted_task_id_array
journal: journal_record_or_null
evidence: selected_evidence_record_or_null
refusal: refusal_record_or_null
journal_record:
keyset: [operation_id, state, path]
state_enum: [committed, rolled_back, prepared, task_writes_started, recovery_required, rollback_in_progress, contended]
constraints: [operation_id equals result operation_id, path is namespaced configured-root transaction journal, additional fields forbidden]
evidence_records:
committed: [actor, occurred_at, validation_attempt_id, checks]
replayed: [source_operation_id, replay_key, checks]
rolled_back: [actor, occurred_at, validation_attempt_id, checks]
recovery_required: [classification, observed_operation_ids, applied_prefix, next_action]
contended: [classification, observed_operation_ids, applied_prefix, next_action]
status: [flows, current, ready, blocked, conflicts]
constraints: [selected record has exactly its listed keys, checks and ids preserve canonical order, additional fields forbidden]
refusal_record:
keyset: [code, stage, field, predicate, path, expected, observed, message]
nullability: {code: required, stage: required, field: nullable, predicate: nullable, path: nullable, expected: nullable, observed: nullable, message: required}
constraints: [additional fields forbidden, code and stage use canonical refusal identifiers, message non-empty]
variants:
committed:
outcome: committed
operation: mutating_operation
journal_state: committed
evidence_schema: committed
nullability: {operation: required, flow_id: required, operation_id: required, plan_revision: required, plan_commit: nullable, state_revision: required, targets: required, journal: required, evidence: required, refusal: null}
replayed:
outcome: replayed
operation: replayable_operation
journal_state: committed
evidence_schema: replayed
nullability: {operation: required, flow_id: required, operation_id: required, plan_revision: required, plan_commit: nullable, state_revision: required, targets: required, journal: required, evidence: required, refusal: null}
rolled_back:
outcome: rolled_back
operation: original_mutating_operation
journal_state: rolled_back
evidence_schema: rolled_back
nullability: {operation: required, flow_id: required, operation_id: required, plan_revision: required, plan_commit: nullable, state_revision: required, targets: required, journal: required, evidence: required, refusal: null}
recovery_required:
outcome: recovery_required
operation: original_mutating_operation
journal_state: prepared_or_task_writes_started_or_recovery_required_or_rollback_in_progress
evidence_schema: recovery_required
nullability: {operation: required, flow_id: required, operation_id: required, plan_revision: required, plan_commit: nullable, state_revision: required, targets: required, journal: required, evidence: required, refusal: null}
contended:
outcome: contended
operation: original_mutating_operation
journal_state: contended
evidence_schema: contended
nullability: {operation: required, flow_id: required, operation_id: required, plan_revision: required, plan_commit: nullable, state_revision: required, targets: required, journal: required, evidence: required, refusal: null}
refused:
outcome: refused
operation: requested_operation_or_null_when_unparseable
journal_state: existing_journal_or_null
evidence_schema: null
nullability: {operation: nullable, flow_id: nullable, operation_id: nullable, plan_revision: nullable, plan_commit: nullable, state_revision: nullable, targets: required, journal: nullable, evidence: null, refusal: required}
status:
outcome: status
operation: status
journal_state: null
evidence_schema: status
nullability: {operation: required, flow_id: nullable, operation_id: null, plan_revision: null, plan_commit: null, state_revision: null, targets: required, journal: null, evidence: required, refusal: null}
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago Changed · +2 lines b566074a68b3
- 1mo ago Changed · +6 lines ee8f3d3e99a4
- 2mo ago First seen · 173 lines · 30 tokens per session scan A 2b142da32054
flow-state is a skill published in the GitHub repository cofin/flow (19 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 30 tokens to every session and 2,392 once invoked, about $0.0001 per session on Opus 5.5. 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.
Other skills, from other repositories
project-init
Use when beads/Dolt database initialization fails, when bd commands return errors about missing databases, when setting up beads in a new project, or when recovering from diverged Dolt history. Handles fresh init, bootstrap from remote, and recovery workflows.
360-feedback-system
360 feedback register: reviewer, subject, review cycle, visibility, due date and score, as CSV, SQL, JSON Schema or Notion on request. Use for 360 reviews or peer feedback cycles.
alumni-re-hire-tracker
Alumni and re-hire register: former role, last working day, re-hire eligibility, current employer and re-engagement date, as CSV, SQL, JSON Schema or Notion on request. Use for alumni outreach.
announcement-board
Announcement board: author, category, department, priority, audience, publish and expiry dates, status and acknowledgements, as CSV, SQL, JSON Schema or Notion on request. Use for internal notices.
stale-sweep
Sweep the googleapis/mcp-toolbox repo for issues and PRs with no real activity in N days (default 60), sort each by whose silence it is (the author's, ours, or nobody's), and draft the nudge or close comment. Use whenever a maintainer asks for a stale sweep, backlog cleanup, or an SLO check, e.g. "stale sweep", "find…
record-gotchas
Capture surprises, workarounds, and rough edges hit while consuming the public surface of Prisma 8, Prisma Compute, or Prisma Postgres — anything a real user of these products would experience. Fires whenever an operator (or agent) writes a workaround, hits a surprising failure mode, or finds undocumented behaviour…