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/packmindhub/packmind/feature-flags-auditnpx skills add PackmindHub/packmind --skill feature-flags-auditgit clone --depth 1 https://github.com/PackmindHub/packmindWhat 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 | $0.00093 | $0.03596 |
| Opus 5 | $0.00046 | $0.01798 |
| Sonnet 5 | $0.00019 | $0.00719 |
| Haiku 4.5 | $0.00009 | $0.00360 |
Grade A, and why
feature-flags-audit 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 3d 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 — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Feature Flags Audit
Produce a synthetic inventory of every feature flag declared in the Packmind codebase, with audience targeting, inferred functional descriptions, and orphan detection. The result is a single markdown table the user can scan at a glance.
Context
Packmind uses a custom feature flag system shared by both frontend and backend. There is no third-party library (LaunchDarkly, GrowthBook, PostHog, etc.). Flags are gated by email domain or exact email via a single central registry. There is no org-based, plan-based, or percentage-rollout mechanism — audience targeting is email-based only. One runtime override does exist: an FF_* env kill-switch (see below) can force a flag fully on or off regardless of the email rule.
Canonical source of truth:
packages/feature-flags/src/registry.ts (the shared, browser-safe @packmind/feature-flags package)
This file exports:
- Flag key constants ending in
_FEATURE_KEY(e.g.SPACES_MANAGEMENT_FEATURE_KEY = 'spaces-management'). - The
FeatureFlagKeyunion of all valid flag string values. DEFAULT_FEATURE_DOMAIN_MAP: aRecord<string, readonly string[]>mapping each flag's string value to a list of allowed entries. Entries starting with@are domain rules (e.g.@packmind.commatches any user at that domain). Entries containing@but not starting with@are exact email matches.- The
isFeatureFlagEnabled()pure function used at call sites (defined inpackages/feature-flags/src/isFeatureFlagEnabled.ts).
The registry is consumed on both sides: the frontend imports it via @packmind/ui (which re-exports the keys, DEFAULT_FEATURE_DOMAIN_MAP, and isFeatureFlagEnabled for back-compat; packages/ui/src/lib/components/content/PMFeatureFlag/PMFeatureFlag.tsx now holds only the PMFeatureFlag React wrapper component). The backend evaluates the same registry through the isFeatureEnabled(flag, { userEmail }) helper in @packmind/node-utils (packages/node-utils/src/featureFlags/isFeatureEnabled.ts), which layers the FF_* env kill-switch on top of the shared email-domain rule.
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.
- 3d ago First seen · 159 lines · 93 tokens per session scan A 06ef9be668c4
feature-flags-audit is a skill published in the GitHub repository PackmindHub/packmind (306 stars, last pushed 5d ago), licensed Apache-2.0. It adds 93 tokens to every session and 3,596 once invoked, about $0.0005 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.
Other skills, from other repositories
token-saver-config
Configure and diagnose token-saver compression settings. Use when the user asks about adjusting compression levels, checking processor status, debugging hook issues, or reviewing savings statistics.
sdd
Execute the Liatrio Spec-Driven Development (SDD) workflow when explicitly invoked by the user. NOTE: this skill is NOT intended to be dynamically loaded or automatically triggered; it should only ever be explicitly called by the user.
loadout-import-workflow
Import another repo's command/skill suite into loadout as a workflow. Use when the user points at an agent framework, plugin, or command pack (e.g. a GitHub repo, a .claude/commands/ directory, a skills pack) and wants its process turned into a loadout [[workflows]] entry — "import this workflow", "turn this plugin…
loadout-migrate
Migrate existing global AI agent instructions into loadout. Use when adopting loadout, importing a CLAUDE.md / AGENTS.md / global agent rules into loadout, setting up /.config/loadout/config.toml, or turning prose agent instructions into reusable fragments and stack-targeted profiles.
loadout-remember
Save a durable, cross-project user preference into loadout's global fragments/profiles — or update the fragment it contradicts. Use when the user states lasting guidance that should follow them across projects and agents ("always X", "never Y", "stop doing Z"), especially when it conflicts with guidance the loadout…
orchestrator
Use this skill when running as an ephemeral orchestrator subagent for a Taniwha project. The skill instructs the agent to read project state from disk, decide the single next action the dispatcher should take, write that decision back to disk, and exit. Trigger this whenever the dispatcher has invoked an orchestrator…