cpi-security

cpi-security is a skill for Claude Code, Codex from PlamenTSV/plamen. It costs 28 tokens per session (1,341 once invoked), scanned A, original, MIT.

A security review guide for cross-program invocations (CPIs), which let one Solana program call another. It checks the called program, forwarded accounts, and signer permissions.

In plain words
What is it for?
It helps inventory CPI calls, verify program IDs, trace forwarded signers, and identify related security findings.
Why use it?
It helps detect cases where user input could replace the intended program or gain unsafe signer privileges.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit It helps inventory CPI calls, verify program IDs, trace forwarded signers, and identify related security findings.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/plamentsv/plamen/cpi-security
Install

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.

Any agent
npx skills add PlamenTSV/plamen --skill cpi-security
Clone the repo
git clone --depth 1 https://github.com/PlamenTSV/plamen

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for cpi-security

README.md
[![agentmods](https://agentmods.dev/badge/skills/plamentsv/plamen/cpi-security/github.svg)](https://agentmods.dev/skills/plamentsv/plamen/cpi-security)
Your own site
<a href="https://agentmods.dev/skills/plamentsv/plamen/cpi-security"><img src="https://agentmods.dev/badge/skills/plamentsv/plamen/cpi-security/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.

agentmods 80×15 button for cpi-security

Your own site · 80×15
<a href="https://agentmods.dev/skills/plamentsv/plamen/cpi-security"><img src="https://agentmods.dev/badge/skills/plamentsv/plamen/cpi-security.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 28 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,341 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00028 $0.01341
Opus 5 $0.00014 $0.00671
Sonnet 5 $0.00006 $0.00268
Haiku 4.5 $0.00003 $0.00134

Measured 7d ago against content hash 18f52437eb88, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

cpi-security 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 7d 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.

agents/skills/solana/cpi-security/SKILL.md · 114 lines

How it starts

The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CPI_SECURITY Skill

Trigger Pattern: CPI flag detected (invoke/invoke_signed/CpiContext usage) Inject Into: Breadth agents, depth agents Finding prefix: [CPI-N] Rules referenced: S3, S5, R1, R4

For every Cross-Program Invocation (CPI) in the Solana program:

1. CPI Target Inventory

Enumerate ALL CPI calls:

# Location CPI Type Target Program Target Instruction Accounts Forwarded Signers Forwarded
1 {file:line} invoke/invoke_signed/CpiContext {program_id} {instruction} {list} {list}

2. Target Program Validation

For each CPI call, verify the target program ID:

CPI Program ID Source Hardcoded/Validated? Spoofable?
{cpi} {where program_id comes from} Hardcoded constant / Validated against known / FROM USER INPUT {if from input: CRITICAL finding}

Attack (S3): If program ID comes from user input without validation, attacker substitutes a malicious program that mimics the expected interface but steals funds. Defense: Always validate program ID against a hardcoded constant or well-known program address.

3. Signer Privilege Tracing

For each CPI that forwards signers:

CPI Signer Source Privilege Level Should Forward? Over-Privileged?
{cpi} {PDA / user wallet / authority} {what the signer can do in target} YES/NO {if forwarding more privilege than needed}

Attack: Forwarding user's wallet as signer to a CPI target that uses it for unintended operations. Pattern: PDA signers via invoke_signed are safe (program-controlled). User wallet forwarding needs careful scoping.

4. Account Reload Audit (S5 - CRITICAL)

For each CPI call, check what accounts are read AFTER the CPI returns:

CPI Accounts Modified by Target Read After CPI? reload() Called? Owner Re-checked?
{cpi} {list accounts CPI can modify} YES/NO YES/NO YES/NO

Read the full file on GitHub · 114 lines

Changes

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.

  1. 7d ago First seen · 114 lines · 28 tokens per session scan A 18f52437eb88

Subscribe to this mod's changes

cpi-security is a skill published in the GitHub repository PlamenTSV/plamen (295 stars, last pushed 2d ago), licensed MIT. It adds 28 tokens to every session and 1,341 once invoked, about $0.0001 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 2026-09-03.