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 Abhillashjadhav/AI-PM-essential-skills --skill pm-context-systemgit clone --depth 1 https://github.com/Abhillashjadhav/AI-PM-essential-skillsWrote 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/abhillashjadhav/ai-pm-essential-skills/pm-context-system)<a href="https://agentmods.dev/skills/abhillashjadhav/ai-pm-essential-skills/pm-context-system"><img src="https://agentmods.dev/badge/skills/abhillashjadhav/ai-pm-essential-skills/pm-context-system/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/abhillashjadhav/ai-pm-essential-skills/pm-context-system"><img src="https://agentmods.dev/badge/skills/abhillashjadhav/ai-pm-essential-skills/pm-context-system.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00198 | $0.00996 |
| Opus 5 | $0.00099 | $0.00498 |
| Sonnet 5 | $0.00040 | $0.00199 |
| Haiku 4.5 | $0.00020 | $0.00100 |
Grade A, and why
pm-context-system 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 11d 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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PM Context System
A maintained memory structure, not a filing cabinet. Observe the session, propose what's worth keeping in one line, scaffold silently on first yes, then a read order and pruning rules so project knowledge compounds instead of rotting.
Step 1 — Observe and propose (no interview)
Don't front-load an interview. Instead, watch the session as it unfolds. The moment a decision gets made, a stakeholder fact surfaces, or project state changes in a way worth remembering, propose exactly one line:
Log to memory: "<decision or fact + why>" — yes/edit/skip
- yes → write it (scaffolding
context/silently first if it doesn't exist yet — see Step 2). - edit → take the user's correction, then write.
- skip → drop it, don't ask again about that same fact.
Never write anything without one of these three responses. Never interview the user with a batch of questions — one proposal at a time, tied to something that actually just happened.
Step 2 — Scaffold (silent, triggered by first yes)
If context/ doesn't exist in the project root yet, create it silently the first time a proposal is approved — no announcement, no empty templates shown up front:
context/
├── INDEX.md ← read-order rules + one-line summary of each file
├── STAKEHOLDERS.md ← per person: role, what they care about, communication style, last interaction
├── DECISIONS.md ← append-only: date, decision, why, alternatives rejected
└── STATE.md ← current project status: active work, blockers, next milestones
Write the approved entry into whichever file it belongs in. Add a pointer in the project's CLAUDE.md: "At session start, read context/INDEX.md first."
Step 3 — Session-start read order
- INDEX.md (cheap, routes everything)
- STATE.md (what's live right now)
- STAKEHOLDERS.md / DECISIONS.md only when the task touches a person or reopens a past decision — progressive disclosure, don't bulk-load.
Step 4 — Session-end catch-all sweep
Real-time proposals (Step 1) cover most of what's worth logging as it happens. Before the session ends (or when the user says "update context"), do one final sweep for anything not already proposed and answered — propose it the same one-line way, never silently write:
- STATE.md: what changed, what's newly blocked/unblocked
- DECISIONS.md: any decision made this session not already logged (date + why + rejected alternatives)
- STAKEHOLDERS.md: only if new information about a person surfaced and wasn't already logged User approves each, then write.
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.
- 11d ago First seen · 65 lines · 198 tokens per session scan A 86954e332ae9
pm-context-system is a skill published in the GitHub repository Abhillashjadhav/AI-PM-essential-skills (3 stars, last pushed 10d ago), licensed MIT. It adds 198 tokens to every session and 996 once invoked, about $0.0010 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 skills, from other repositories
session-state
Track implementation decisions and progress in specs/state.yaml to prevent context rot. Use at the start of a session to load context, and whenever a significant decision is made or a milestone is reached.
proactive-memory
Turn repeated failures, corrections, and recurring requests into durable follow-up work.
kb
Use when querying and maintaining the knowledge base for project context, decisions, and architecture documentation on session start.
berkahkarya-orchestrator
Use when orchestrate multi-skill workflows by routing tasks to the right agents and coordinating cross-platform operations.
vilona-activate
Use when activate an AI general manager persona with full context awareness and multi-user adaptation.
teamwork
Use when dynamically creates and manages AI agent teams for complex tasks. Invoke when user requests multi-agent collaboration, complex project execution, or when tasks require specialized roles and coordinated workflow. Use when working with teamwork.