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/adit-jain-srm/skill-forge/context-buildernpx skills add Adit-Jain-srm/skill-forge --skill context-buildergit clone --depth 1 https://github.com/Adit-Jain-srm/skill-forgeWrote 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/adit-jain-srm/skill-forge/context-builder)<a href="https://agentmods.dev/skills/adit-jain-srm/skill-forge/context-builder"><img src="https://agentmods.dev/badge/skills/adit-jain-srm/skill-forge/context-builder.svg" alt="Measured on agentmods" 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 | $0.00105 | $0.01075 |
| Opus 5 | $0.00053 | $0.00537 |
| Sonnet 5 | $0.00021 | $0.00215 |
| Haiku 4.5 | $0.00011 | $0.00108 |
Grade A, and why
context-builder 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 4d 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 — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Context Builder
Overview
Create a shared vocabulary between you and the agent. Once built, the agent speaks YOUR language — concise, precise, domain-native.
The Problem This Solves
Without shared context:
- Agent uses 20 words where your domain has 1 term
- Variables named generically ("data", "handler", "service") instead of domain terms
- Every session starts from scratch explaining the same concepts
- Conversations are verbose because the agent doesn't know your jargon
With CONTEXT.md:
- Agent speaks your domain language natively
- "the materialization cascade" replaces "when a lesson inside a section of a course is made real"
- Code uses domain terms for variables, functions, modules
- Sessions start with shared understanding already loaded
Process
1. Interview (grill the user)
Ask these questions ONE AT A TIME:
- "What is this project? One sentence."
- "What are the 3-5 most important CONCEPTS in this domain?"
- For each concept: "Define it in one sentence. What terms should I AVOID using for this?"
- "Are there relationships between these concepts? (X contains many Y, Y belongs to one X)"
- "Any terms that were previously confusing or ambiguous? What did you resolve them to?"
- "What actions/verbs are specific to this domain?" (e.g., "materialize", "triage", "hydrate")
2. Generate CONTEXT.md
Write to the project root:
# [Project Name]
[One-sentence description]
## Language
**[Term 1]**:
[Definition in one sentence]
_Avoid_: [terms NOT to use for this concept]
**[Term 2]**:
[Definition]
_Avoid_: [alternatives to avoid]
## Relationships
- A **[Term 1]** contains many **[Term 2]**s
- A **[Term 2]** belongs to one **[Term 1]**
## Flagged Ambiguities
- "[confusing term]" was previously used for both X and Y — resolved: [how it's now used]
3. Persistence
Once CONTEXT.md exists:
- Agent reads it at session start
- Uses defined terms in ALL communication
- Names variables/functions using the vocabulary
- Flags when new undefined terms appear: "Should I add '[term]' to CONTEXT.md?"
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.
- 4d ago First seen · 127 lines · 105 tokens per session scan A 381ee48122ee
context-builder is a skill published in the GitHub repository Adit-Jain-srm/skill-forge (2 stars, last pushed 2mo ago), licensed MIT. It adds 105 tokens to every session and 1,075 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 2026-08-31.
Other skills, from other repositories
user-research-cookiy
End-to-end user research assistant — qualitative and quantitative. Use this skill whenever the user mentions user research, user interviews, discussion guides, interview guides, research plans, qualitative research, quantitative research, user surveys, survey design, usability studies, participant recruitment…
crossframe-suite
CrossFrame Suite explicit-only router. Use only when the user explicitly names crossframe-suite, $crossframe-suite, /crossframe-suite, or asks to use the CrossFrame suite; do not trigger implicitly for ordinary analysis, writing, public, organization, debate, teaching, reading, or review tasks. After explicit…
crossframe-essay
CrossFrame Essay explicit-only Chinese critical insight writing skill. Use only when the user explicitly names crossframe-essay, $crossframe-essay, /crossframe-essay, or asks to use CrossFrame Essay; do not trigger implicitly for ordinary writing, commentary, article, editing, or long-answer tasks. Suite-directed use…
crossframe-promax
Use when the user explicitly names crossframe-promax, CrossFrame ProMax, $crossframe-promax, or /crossframe-promax; never infer activation from generic requests for depth, completeness, maximum effort, or long output.
crossframe-max
Use when the user explicitly invokes crossframe-max, /crossframe-max, $crossframe-max, maximum CrossFrame mode, exhaustive structural projection, full-scale world-model interpretation, no word-limit complete explanation, or v6 meta-runtime design review for skills, prompts, agents, tools, templates, scripts, and…
crossframe-inquiry
经由 crossframe-suite 完成态接管或用户显式命令使用,不隐式响应。把 CrossFrame 诊断、文章、review、claim ledger 和机制候选转化为结构追问,帮助用户继续深思、反证、补证、迁移应用或收束。.