CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

upgradeable-proxy

Proxy upgrade patterns and their bugs — uninitialized implementation, storage slot collisions, selector clashes, unprotected upgrade auth.

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./packages/decepticon/decepticon/skills/standard/contracts/upgradeable-proxy/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

82%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a strong, actionability-first audit playbook: executable greps, slither commands, runnable Foundry PoCs, and a remediation template with no filler. Minor gaps are slight duplication (Parity told twice), an implicitly rather than explicitly sequenced workflow, and inline PoC/exemplar material that could be split into a references file.

DimensionReasoningScore

Conciseness

The body is dense and code-first, assuming Claude already knows proxy mechanics ('Proxies separate storage (proxy) from logic (implementation)'), matching the 'efficient; minor instances of over-explanation that could be trimmed' anchor. It is not 5 because of small duplications — the Parity incident appears in both section 1 ('Parity multisig 2017 — $280M frozen') and the exemplars ('Parity multisig (Jul + Nov 2017): $30M stolen, $280M frozen'), and prose sometimes restates what code comments already say ('admin set in constructor, proxy reads 0x0'); it is not 3 because nearly all content is domain-specific value (incident details, exact commands) rather than explanation of known concepts.

4 / 5

Actionability

Every section gives executable guidance: grep commands ('grep -rn '_authorizeUpgrade' src/'), tool invocations ('slither-check-upgradeability V1.sol V2.sol --proxy-name MyProxy'), copy-paste-ready Foundry PoC tests ('vm.prank(attacker); proxy.upgradeTo(malicious);'), and complete vulnerable/safe code pairs plus a remediation template. This matches the 'fully executable; copy-paste ready code or commands; specific examples cover the common cases' anchor — it is not 4 because the commands and tests are complete and runnable as written, with no missing key details.

5 / 5

Workflow Clarity

A sequenced audit checklist (sections 1-8) each with its own verification command, followed by PoC templates for confirmation, a CVSS severity stage, and remediation with CI gates ('forge test --match-contract Test_Upgrade', 'slither-check-upgradeability V1 V2 in CI') — matching 'clear sequence with most checkpoints present; minor validation gaps'. It is not 5 because the checklist → PoC → CVSS → remediation progression is implied by layout rather than explicitly stated, and there is no guidance on what to do when a check hits vs. misses; it is not 3 because each check has an explicit verification command and the PoC templates serve as confirmation checkpoints.

4 / 5

Progressive Disclosure

A single-file skill with no bundle directories (references/, scripts/, assets/ are absent) whose content is organized under clear headers ('Audit checklist', 'PoC template (Foundry)', 'CVSS', 'Defender remediation', 'Known exemplars') — matching 'good structure; most content is appropriately placed; minor organization gaps'. It is not 5 because at ~245 lines the PoC templates and incident exemplars are separable material that could live in a references file, keeping some detail inline that a well-split bundle would externalize; it is not 3 because no section is an oversized dump requiring extraction and navigation is easy.

4 / 5

Total

17

/

20

Passed

Description

70%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific and distinctive, naming four concrete proxy-upgrade bug classes with natural trigger terms. Its main weakness is the missing 'Use when...' clause — trigger guidance lives in metadata.when_to_use instead of the description itself, capping completeness at 3.

Suggestions

Append a 'Use when...' clause to the description, e.g.: 'Use when auditing upgradeable proxies (UUPS, Transparent, Diamond/EIP-2535) or when the user mentions proxy upgrades, EIP-1967, or initializer/storage-collision bugs.' This lifts completeness above the cap of 3.

Fold the natural keywords currently only in metadata.when_to_use ('upgradeable proxy', 'UUPS', 'EIP-1967') into the description text so trigger-term coverage includes the synonyms users actually say.

Mention the three proxy patterns the body covers (Transparent, UUPS, Diamond) in the description so the 'what' fully matches the skill's scope and closes the coverage gap.

DimensionReasoningScore

Specificity

The description lists four concrete bug classes — "uninitialized implementation, storage slot collisions, selector clashes, unprotected upgrade auth" — matching the anchor 'lists several specific actions; minor gaps in coverage'. It is not 5 because coverage has gaps (Diamond/EIP-2535 facet bugs and constructor-vs-initializer issues covered in the body are absent) and the items are named domains rather than actions; it is not 3 because four specific concrete items exceed '1-2 concrete actions'.

4 / 5

Completeness

The 'what' is clear ("Proxy upgrade patterns and their bugs" plus four named bug classes), but the description contains no 'Use when...' clause or equivalent explicit trigger guidance, which caps completeness at 3 per the judging guidelines. It is not 4 because the 'when' is entirely absent from the description rather than merely imprecise; it is not 2 because the 'what' is concrete and specific rather than vague.

3 / 5

Trigger Term Quality

Phrases like "proxy upgrade", "storage slot collisions", "selector clashes", and "uninitialized implementation" are natural terms a smart-contract auditor would say, matching the 'good keyword coverage; a few natural terms missing' anchor. It falls short of 5 because common synonyms/extensions a user would naturally say — "UUPS", "EIP-1967", "transparent proxy", "diamond", "upgradeable" — appear only in metadata.when_to_use, not in the description field itself.

4 / 5

Distinctiveness Conflict Risk

"Proxy upgrade patterns and their bugs — uninitialized implementation, storage slot collisions, selector clashes, unprotected upgrade auth" carves out a clear niche with triggers specific to upgradeable proxies, matching the 'clear niche with distinct triggers; minimal conflict risk' anchor. It is not 4 because the named bug classes (selector clashes, storage slot collisions) are unique to the proxy-upgrade domain and would not plausibly fire a general access-control or reentrancy skill.

5 / 5

Total

16

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

Total

14

/

16

Passed

Repository

PurpleAILAB/Decepticon

Path

packages/decepticon/decepticon/skills/standard/contracts/upgradeable-proxy/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.