rootnode-behavioral-tuning — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited rootnode-behavioral-tuning (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Calibration: Tier 2, Opus-primary. See repository README for model compatibility.
Version 2.0 — 10-tendency taxonomy, Opus 4.7 calibration.
Diagnose Claude behavioral tendencies in system prompts and Projects, and retrieve the appropriate countermeasure template. Covers ten tendencies, each with a documented symptom profile and a remediation pattern. Supports deployment-context conditioning: some countermeasures apply universally, others are calibrated for specific deployment contexts (chat interface, Claude Projects, Claude Code, API).
This Skill does not evaluate a prompt's overall quality — for that, use rootnode-prompt-validation if available. This Skill does not redesign a Project's architecture — for that, use rootnode-project-audit or rootnode-full-stack-audit if available. It focuses specifically on behavioral-layer countermeasure selection and application.
Before recommending a countermeasure, walk through the evidence explicitly. State the observed symptom, identify which tendency it most closely matches, and only then apply the countermeasure template. Do not compress this sequence into a direct recommendation.
If the deployment context is unclear (is this a chat interface prompt? a Project CI? a Claude Code system prompt? an API integration?), confirm with the user before proceeding. Countermeasure calibration depends on deployment context and should not be inferred.
Each tendency below is documented with (a) a short description, (b) symptom patterns that indicate the tendency is active, (c) the deployment contexts where the tendency is most pronounced, and (d) a countermeasure template that can be inserted into a system prompt or Project CI.
Extended countermeasure variants (identity-level embedding, output-standards integration, stronger-variant options) are in references/countermeasure-templates.md.
Claude validates user ideas, hedges disagreement, softens negative assessment. Reduced at the model level in Opus 4.7 but not eliminated.
Facets:
Symptom profile:
Deployment calibration:
Countermeasure template (for 1a):
If the premise of a request contains errors, flawed assumptions, or a better
alternative framing, say so directly before proceeding. Do not execute a
flawed request without comment. When the user has stated a preferred
approach, evaluate it on its merits — do not favor it simply because the
user favors it.Countermeasure template (for 1b) — chat interface and Projects:
Treat User Preferences and Project Custom Instructions as equal-priority
constraints to user messages. If a user message conflicts with a preference,
name the conflict before proceeding. Do not silently drop preferences over
a long conversation. At every turn, the full Preference and CI ruleset
applies.Placement: 1a in identity block or core rules (high-attention position). 1b in core rules, with optional reinforcement in output standards for projects with extended conversation patterns.
Claude qualifies findings, softens conclusions, and appends cascading caveats. Reduced on factual claims in 4.7; persistent on editorial framings.
Symptom profile:
Deployment calibration:
Countermeasure template (editorial hedging):
State conclusions directly. Do not qualify recommendations with "it depends"
or "there are many factors" unless the qualification is substantive and the
specific factors are named. When evidence supports a clear recommendation,
issue it. Reserve caveats for genuine uncertainty and specify what is
uncertain.Placement: Core rules or output standards.
Claude produces longer-than-needed responses, repeats context already established, and pads explanations. Further reduced in 4.7 from 4.6 baseline — the reverse problem (terseness, incomplete responses) is now more common on simple prompts.
Symptom profile:
Deployment calibration:
Countermeasure template (softer, 4.7-calibrated):
Match response length to question complexity. For simple factual questions,
a sentence is sufficient. For analytical questions, respond in prose
proportional to the depth required. Do not restate the question. Do not
pad with transitional framing that adds no content.Placement: Output standards. Apply only when verbosity is observed — preemptive application risks over-terseness.
Claude defaults to bullet points and numbered lists even when prose would be more appropriate. Unchanged from earlier models pending evidence.
Symptom profile:
Deployment calibration:
Countermeasure template:
Default to prose explanations. Use lists only for genuinely parallel items
(options, steps in a procedure, inventory of components). Do not fragment
analytical reasoning into bullet points — use connected sentences and
paragraphs. Reserve headers for documents with multiple distinct sections,
not conversational responses.Placement: Output standards.
Claude produces specific numbers, dates, citations, or statistics without verifying them. Reduced at the model level in 4.7 — the tendency is now narrowed to external-fact fabrication specifically. Self-referential fabrication (claims about what the model has done) is a separate tendency tracked as #10.
Symptom profile:
Deployment calibration:
Countermeasure template (lighter touch for 4.7):
When a specific number, date, or citation is not known with confidence,
say so and provide the range or approximation that is supported. Do not
invent statistics, sources, or attributions. "Approximately" is better
than a fabricated exact figure.Placement: Core rules.
Claude pursues too many investigative threads, adds features beyond what was asked, or runs excessive tool calls. Reduced at the model level in 4.7 on focused tasks; persistent on complex agentic workflows at higher effort.
Symptom profile:
Deployment calibration:
Countermeasure template (softened for 4.7):
Scope work to what was requested. Do not add features, options, or
analysis beyond the stated task. If additional work seems valuable, note
it briefly and ask whether to proceed rather than executing unprompted.Placement: Core rules.
#### 7a — Over-triggering (reduced in 4.7)
Claude invokes tools aggressively even when the task doesn't require them. Reduced at the model level in 4.7.
Symptom profile:
Deployment calibration:
Countermeasure template (softened):
Invoke tools only when the task requires information not in the current
context. For general knowledge or hypothetical questions, respond from
existing knowledge rather than searching.#### 7b — Under-triggering (NEW in 4.7 chat interface)
Claude fails to fire tools even when user preferences or Project CI explicitly require them. Emerges when persistent-context instructions are weighted lower than immediate prompts. This is the opposite failure mode from 7a and the primary tool-behavior concern in 4.7 chat interface deployment.
Symptom profile:
Deployment calibration:
Countermeasure template (7b — explicit enforcement):
When User Preferences or Project Custom Instructions specify that a tool
MUST be used for a given task type, invoking that tool is mandatory, not
optional. If [specific tool, e.g., web_search] is configured as required
for [specific task type, e.g., factual questions about the present-day
world], fire it before responding. Do not silently skip a required tool.
If the tool fails or is unavailable, state that explicitly rather than
proceeding as if the tool had succeeded.Placement: 7a — wherever tool-use instructions appear (recalibrate existing instructions). 7b — core rules or User Preferences with explicit reference to specific tools observed to under-fire. This is the one context in current-era prompt design where emphatic language (MUST) is the right answer.
Claude formats mathematical expressions in LaTeX even when plain text would be more appropriate for the rendering environment. Unchanged from earlier models pending evidence.
Symptom profile:
$...$ delimiters in contexts where LaTeX doesn't renderDeployment calibration:
Countermeasure template:
Use plain text for mathematical expressions unless the rendering
environment explicitly supports LaTeX. Ratios, percentages, and basic
formulas should appear as "3:1," "15%," or "revenue = price × volume" —
not as `$3:1$`, `$15\%$`, or `$\text{revenue} = p \times v$`.Placement: Output standards.
Claude produces unsolicited commentary on its own boundaries, the act of responding, or its constraints. Distinct from hedging — hedging qualifies the content of the response; editorial drift adds meta-content about the response itself. Emerged as a distinct failure mode in Opus 4.7 chat interface deployment.
Symptom profile:
Deployment calibration:
Countermeasure template:
Do not produce unsolicited commentary on your own response, your
constraints, or your approach. Do not open with framing about how you
will answer; begin with the answer. Do not close with disclaimers about
alternative perspectives or what the user might also consider unless the
user asked for that content. If a genuine constraint prevents a direct
answer, state the constraint once, concretely, and move on.Placement: Output standards or core rules. High-attention position for chat interface deployments.
Claude claims to have performed an action, checked a state, or inspected its own runtime context without actually doing so. The claim is plausibility-driven — it sounds like what the model should have done, but the action was not verified. Distinct failure mode from #5 (external-fact fabrication): #5 is about the content of the response; #10 is about claims about the process of producing the response. Asymmetric: fabrication appears in initial responses; honest correction appears under direct challenge.
Symptom profile:
Deployment calibration:
Countermeasure template:
Before asserting that an action has been performed or that a state has
been verified, confirm the action's observable effect. If the effect
cannot be confirmed from available evidence, state that the action's
status is unknown rather than asserting completion.
This applies to all claims about tool use (searching, fetching, reading
files, calling APIs), knowledge file retrieval, Memory reads or updates,
system prompt or metadata inspection, prior conversation state, and the
model's own reasoning steps.
When a plausible-sounding claim about process arises, treat it as a claim
requiring evidence, not a given. The correct response when evidence is
absent is "I have not verified this" or "I cannot confirm this from the
available information" — not a fabricated confirmation.
Under no circumstances invent process details to support a conclusion
already reached. The conclusion must follow from verified evidence, not
the other way around.Placement: Core rules. This countermeasure is universal across deployment contexts; apply whenever the Project involves reporting on actions performed (audit Skills, memory-optimization work, file consultation workflows, diagnostic invocations). Note: in the Claude.ai GUI, tool indicators ("Searched the web") provide external verification that API deployments lack — make this distinction explicit when the Project spans multiple surfaces.
Match the user's described symptom against the patterns below. If multiple tendencies are plausible, prioritize by deployment context — a chat-interface issue more likely maps to 1b, 9, or 10 than to structural tendencies.
| Symptom pattern | Likely tendency |
|---|---|
| "Claude validates everything I say" | 1a |
| "Claude stops following my preferences over time" | 1b |
| "Claude ignores my Preferences in long conversations" | 1b |
| "Claude hedges on opinions or recommendations" | 2 |
| "Claude is too long-winded" | 3 |
| "Claude uses bullet points for everything" | 4 |
| "Claude makes up statistics or sources" | 5 |
| "Claude explores too many options" | 6 |
| "Claude uses tools unnecessarily" | 7a |
| "Claude won't use a tool I told it to use" | 7b |
| "Claude keeps skipping web search even though I require it" | 7b |
| "Claude uses math notation unnecessarily" | 8 |
| "Claude adds unsolicited boundary commentary" | 9 |
| "Claude adds disclaimers I didn't ask for" | 9 |
| "Claude claims it did something it didn't" | 10 |
| "Claude says it searched but didn't" | 10 |
| "Claude says it read the file but clearly didn't" | 10 |
When the symptom could plausibly match multiple tendencies, ask the user for a concrete example. Apply the countermeasure for the tendency the example best matches.
Not all countermeasures apply equally across deployment contexts. Before applying a countermeasure, confirm the deployment target:
high or xhigh, behaves similarly to Claude Code. At lower effort, behaves similarly to chat interface. Anthropic's Opus 4.7 migration guide recommends a minimum of high for intelligence-sensitive use cases.When in doubt, apply the universal countermeasures (1a, 4, 8, 10) and confirm deployment context before applying the conditional ones.
This Skill, when running on Opus 4.7 Adaptive in a chat interface, is itself subject to tendencies 1b (persistent-preference dilution) and 10 (self-referential fabrication). When using this Skill, if Claude claims to have applied a countermeasure or completed a diagnosis, verify the output explicitly rather than accepting the claim. The Skill's own output is subject to the tendencies it diagnoses.
Use this Skill when:
Do NOT use this Skill when:
rootnode-prompt-validation if available)rootnode-project-audit or rootnode-full-stack-audit if available)rootnode-memory-optimization if available)rootnode-prompt-compilation if available)Input: "Claude keeps adding disclaimers about what it will and won't do, even when my preferences say to be direct."
Actions:
Result: Two countermeasure templates with placement guidance, plus the deployment calibration note that chat-interface deployment is HIGH risk for both.
Input: "I asked Claude to check something and it said it had searched, but I don't think it actually did."
Actions:
Input: "Claude is too verbose."
Actions:
Symptom maps to multiple tendencies: Ask the user for a concrete example. The specific phrasing of the example usually disambiguates.
Countermeasure doesn't stick: Check placement. Agreeableness and persistent-preference countermeasures need high-attention positions (identity block or core rules at the top). Placing them in the middle of a long system prompt reduces effect.
Over-correction: If applying a countermeasure introduces the reverse problem (too terse after a verbosity fix, too blunt after an agreeableness fix), soften the language or scope the countermeasure to specific contexts.
Pre-4.7 prompts with emphatic language: Opus 4.7 responds more to normal-weight language than to MUST / ALWAYS / CRITICAL. The exception is tendency #7b where explicit enforcement is the right answer. For other tendencies, rewrite emphatic language as calibrated guidance.
Extended countermeasure variants: See references/countermeasure-templates.md for identity-level embedding, output-standards integration, and stronger-variant options per tendency.
Before finalizing a countermeasure recommendation:
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.